Weiwei Wan

Hao Chen, Weiwei Wan, Masaki Matsushita et al.

Deep learning methods have recently exhibited impressive performance in object detection. However, such methods needed much training data to achieve high recognition accuracy, which was time-consuming and required considerable manual work like labeling images. In this paper, we automatically prepare training data using robots. Considering the low efficiency and high energy consumption in robot motion, we proposed combining robotic in-hand observation and data synthesis to enlarge the limited data set collected by the robot. We first used a robot with a depth sensor to collect images of objects held in the robot's hands and segment the object pictures. Then, we used a copy-paste method to synthesize the segmented objects with rack backgrounds. The collected and synthetic images are combined to train a deep detection neural network. We conducted experiments to compare YOLOv5x detectors trained with images collected using the proposed method and several other methods. The results showed that combined observation and synthetic images led to comparable performance to manual data preparation. They provided a good guide on optimizing data configurations and parameter settings for training detectors. The proposed method required only a single process and was a low-cost way to produce the combined data. Interested readers may find the data sets and trained models from the following GitHub repository: github.com/wrslab/tubedet

Hao Chen, Weiwei Wan, Masaki Matsushita et al.

Accurate robotic manipulation of test tubes in biology and medical industries is becoming increasingly important to address workforce shortages and improve worker safety. The detection and localization of test tubes are essential for the robots to successfully manipulate test tubes. In this paper, we present a framework to detect and estimate poses for the in-rack test tubes using color and depth data. The methodology involves the utilization of a YOLO object detector to effectively classify and localize both the test tubes and the tube racks within the provided image data. Subsequently, the pose of the tube rack is estimated through point cloud registration techniques. During the process of estimating the poses of the test tubes, we capitalize on constraints derived from the arrangement of rack slots. By employing an optimization-based algorithm, we effectively evaluate and refine the pose of the test tubes. This strategic approach ensures the robustness of pose estimation, even when confronted with noisy and incomplete point cloud data.

Vitor Hideyo Isume, Takuya Kiyokawa, Natsuki Yamanobe et al.

Inspired by traditional handmade crafts, where a person improvises assemblies based on the available objects, we formally introduce the Craft Assembly Task. It is a robotic assembly task that involves building an accurate representation of a given target object using the available objects, which do not directly correspond to its parts. In this work, we focus on selecting the subset of available objects for the final craft, when the given input is an RGB image of the target in the wild. We use a mask segmentation neural network to identify visible parts, followed by retrieving labelled template meshes. These meshes undergo pose optimization to determine the most suitable template. Then, we propose to simplify the parts of the transformed template mesh to primitive shapes like cuboids or cylinders. Finally, we design a search algorithm to find correspondences in the scene based on local and global proportions. We develop baselines for comparison that consider all possible combinations, and choose the highest scoring combination for common metrics used in foreground maps and mask accuracy. Our approach achieves comparable results to the baselines for two different scenes, and we show qualitative results for an implementation in a real-world scenario.

Vitor Hideyo Isume, Takuya Kiyokawa, Natsuki Yamanobe et al.

Inspired by traditional handmade crafts, where a person improvises assemblies based on the available objects, we formally introduce the Craft Assembly Task. It is a robotic assembly task that involves building an accurate representation of a given target object using the available objects, which do not directly correspond to its parts. In this work, we focus on selecting the subset of available objects for the final craft, when the given input is an RGB image of the target in the wild. We use a mask segmentation neural network to identify visible parts, followed by retrieving labeled template meshes. These meshes undergo pose optimization to determine the most suitable template. Then, we propose to simplify the parts of the transformed template mesh to primitive shapes like cuboids or cylinders. Finally, we design a search algorithm to find correspondences in the scene based on local and global proportions. We develop baselines for comparison that consider all possible combinations, and choose the highest scoring combination for common metrics used in foreground maps and mask accuracy. Our approach achieves comparable results to the baselines for two different scenes, and we show qualitative results for an implementation in a real-world scenario.

Rewida Ali, Cristian C. Beltran-Hernandez, Weiwei Wan et al.

Human-robot cooperation is essential in environments such as warehouses and retail stores, where workers frequently handle deformable objects like paper, bags, and fabrics. Coordinating robotic actions with human assistance remains difficult due to the unpredictable dynamics of deformable materials and the need for adaptive force control. To explore this challenge, we focus on the task of gift wrapping, which exemplifies a long-horizon manipulation problem involving precise folding, controlled creasing, and secure fixation of paper. Success is achieved when the robot completes the sequence to produce a neatly wrapped package with clean folds and no tears. We propose a learning-based framework that integrates a high-level task planner powered by a large language model (LLM) with a low-level hybrid imitation learning (IL) and reinforcement learning (RL) policy. At its core is a Sub-task Aware Robotic Transformer (START) that learns a unified policy from human demonstrations. The key novelty lies in capturing long-range temporal dependencies across the full wrapping sequence within a single model. Unlike vanilla Action Chunking with Transformer (ACT), typically applied to short tasks, our method introduces sub-task IDs that provide explicit temporal grounding. This enables robust performance across the entire wrapping process and supports flexible execution, as the policy learns sub-goals rather than merely replicating motion sequences. Our framework achieves a 97% success rate on real-world wrapping tasks. We show that the unified transformer-based policy reduces the need for specialized models, allows controlled human supervision, and effectively bridges high-level intent with the fine-grained force control required for deformable object manipulation.

Hiroki Hanai, Takuya Kiyokawa, Weiwei Wan et al.

Robotic packaging using wrapping paper poses significant challenges due to the material's complex deformation properties. The packaging process itself involves multiple steps, primarily categorized as folding the paper or creating creases. Small deviations in the robot's arm trajectory or force vector can lead to tearing or wrinkling of the paper, exacerbated by the variability in material properties. This study introduces a novel framework that combines imitation learning and reinforcement learning to enable a robot to perform each step of the packaging process efficiently. The framework allows the robot to follow approximate trajectories of the tool-center point (TCP) based on human demonstrations while optimizing force control parameters to prevent tearing or wrinkling, even with variable wrapping paper materials. The proposed method was validated through ablation studies, which demonstrated successful task completion with a significant reduction in tear and wrinkle rates. Furthermore, the force control strategy proved to be adaptable across different wrapping paper materials and robust against variations in the size of the target object.

Hao Chen, Takuya Kiyokawa, Zhengtao Hu et al.

Grasping unknown objects from a single view has remained a challenging topic in robotics due to the uncertainty of partial observation. Recent advances in large-scale models have led to benchmark solutions such as GraspNet-1Billion. However, such learning-based approaches still face a critical limitation in performance robustness for their sensitivity to sensing noise and environmental changes. To address this bottleneck in achieving highly generalized grasping, we abandon the traditional learning framework and introduce a new perspective: similarity matching, where similar known objects are utilized to guide the grasping of unknown target objects. We newly propose a method that robustly achieves unknown-object grasping from a single viewpoint through three key steps: 1) Leverage the visual features of the observed object to perform similarity matching with an existing database containing various object models, identifying potential candidates with high similarity; 2) Use the candidate models with pre-existing grasping knowledge to plan imitative grasps for the unknown target object; 3) Optimize the grasp quality through a local fine-tuning process. To address the uncertainty caused by partial and noisy observation, we propose a multi-level similarity matching framework that integrates semantic, geometric, and dimensional features for comprehensive evaluation. Especially, we introduce a novel point cloud geometric descriptor, the C-FPFH descriptor, which facilitates accurate similarity assessment between partial point clouds of observed objects and complete point clouds of database models. In addition, we incorporate the use of large language models, introduce the semi-oriented bounding box, and develop a novel point cloud registration approach based on plane detection to enhance matching accuracy under single-view conditions. Videos are available at https://youtu.be/qQDIELMhQmk.

Haoyuan Wang, Zihao Dong, Hongliang Lei et al.

In this work, we conducted research on deformable object manipulation by robots based on demonstration-enhanced reinforcement learning (RL). To improve the learning efficiency of RL, we enhanced the utilization of demonstration data from multiple aspects and proposed the HGCR-DDPG algorithm. It uses a novel high-dimensional fuzzy approach for grasping-point selection, a refined behavior-cloning method to enhance data-driven learning in Rainbow-DDPG, and a sequential policy-learning strategy. Compared to the baseline algorithm (Rainbow-DDPG), our proposed HGCR-DDPG achieved 2.01 times the global average reward and reduced the global average standard deviation to 45% of that of the baseline algorithm. To reduce the human labor cost of demonstration collection, we proposed a low-cost demonstration collection method based on Nonlinear Model Predictive Control (NMPC). Simulation experiment results show that demonstrations collected through NMPC can be used to train HGCR-DDPG, achieving comparable results to those obtained with human demonstrations. To validate the feasibility of our proposed methods in real-world environments, we conducted physical experiments involving deformable object manipulation. We manipulated fabric to perform three tasks: diagonal folding, central axis folding, and flattening. The experimental results demonstrate that our proposed method achieved success rates of 83.3%, 80%, and 100% for these three tasks, respectively, validating the effectiveness of our approach. Compared to current large-model approaches for robot manipulation, the proposed algorithm is lightweight, requires fewer computational resources, and offers task-specific customization and efficient adaptability for specific tasks.

Hao Chen, Takuya Kiyokawa, Weiwei Wan et al.

Robotic pick-and-place has been researched for a long time to cope with uncertainty of novel objects and changeable environments. Past works mainly focus on learning-based methods to achieve high precision. However, they have difficulty being generalized for the limitation of specified training models. To break through this drawback of learning-based approaches, we introduce a new perspective of similarity matching between novel objects and a known database based on category-association to achieve pick-and-place tasks with high accuracy and stabilization. We calculate the category name similarity using word embedding to quantify the semantic similarity between the categories of known models and the target real-world objects. With a similar model identified by a similarity prediction function, we preplan a series of robust grasps and imitate them to plan new grasps on the real-world target object. We also propose a distance-based method to infer the in-hand posture of objects and adjust small rotations to achieve stable placements under uncertainty. Through a real-world robotic pick-and-place experiment with a dozen of in-category and out-of-category novel objects, our method achieved an average success rate of 90.6% and 75.9% respectively, validating the capacity of generalization to diverse objects.

Xinyi Zhang, Yukiyasu Domae, Weiwei Wan et al.

Wire harnesses are essential connecting components in manufacturing industry but are challenging to be automated in industrial tasks such as bin picking. They are long, flexible and tend to get entangled when randomly placed in a bin. This makes it difficult for the robot to grasp a single one in dense clutter. Besides, training or collecting data in simulation is challenging due to the difficulties in modeling the combination of deformable and rigid components for wire harnesses. In this work, instead of directly lifting wire harnesses, we propose to grasp and extract the target following a circle-like trajectory until it is untangled. We learn a policy from real-world data that can infer grasps and separation actions from visual observation. Our policy enables the robot to efficiently pick and separate entangled wire harnesses by maximizing success rates and reducing execution time. To evaluate our policy, we present a set of real-world experiments on picking wire harnesses. Our policy achieves an overall 84.6% success rate compared with 49.2% in baseline. We also evaluate the effectiveness of our policy under different clutter scenarios using unseen types of wire harnesses. Results suggest that our approach is feasible for handling wire harnesses in industrial bin picking.

Xinyi Zhang, Keisuke Koyama, Yukiyasu Domae et al.

This paper addresses the problem of picking up only one object at a time avoiding any entanglement in bin-picking. To cope with a difficult case where the complex-shaped objects are heavily entangled together, we propose a topology-based method that can generate non-tangle grasp positions on a single depth image. The core technique is entanglement map, which is a feature map to measure the entanglement possibilities obtained from the input image. We use the entanglement map to select probable regions containing graspable objects. The optimum grasping pose is detected from the selected regions considering the collision between robot hand and objects. Experimental results show that our analytic method provides a more comprehensive and intuitive observation of entanglement and exceeds previous learning-based work in success rates. Especially, our topology-based method does not rely on any object models or time-consuming training process, so that it can be easily adapted to more complex bin-picking scenes.

Issei Sera, Natsuki Yamanobe, Ixchel G. Ramirez-Alpizar et al.

This paper proposes a robot assembly planning method by automatically reading the graphical instruction manuals design for humans. Essentially, the method generates an Assembly Task Sequence Graph (ATSG) by recognizing a graphical instruction manual. An ATSG is a graph describing the assembly task procedure by detecting types of parts included in the instruction images, completing the missing information automatically, and correcting the detection errors automatically. To build an ATSG, the proposed method first extracts the information of the parts contained in each image of the graphical instruction manual. Then, by using the extracted part information, it estimates the proper work motions and tools for the assembly task. After that, the method builds an ATSG by considering the relationship between the previous and following images, which makes it possible to estimate the undetected parts caused by occlusion using the information of the entire image series. Finally, by collating the total number of each part with the generated ATSG, the excess or deficiency of parts are investigated, and task procedures are removed or added according to those parts. In the experiment section, we build an ATSG using the proposed method to a graphical instruction manual for a chair and demonstrate the action sequences found in the ATSG can be performed by a dual-arm robot execution. The results show the proposed method is effective and simplifies robot teaching in automatic assembly.

Ang Zhang, Keisuke Koyama, Weiwei Wan et al.

Pivoting gait is efficient for manipulating a big and heavy object with relatively small manipulating force, in which a robot iteratively tilts the object, rotates it around the vertex, and then puts it down to the floor. However, pivoting gait can easily fail even with a small external disturbance due to its instability in nature. To cope with this problem, we propose a controller to robustly control the object motion during the pivoting gait by introducing two gait modes, i.e., one is the double-support mode, which can manipulate a relatively light object with faster speed, and the other is the quadruple-support mode, which can manipulate a relatively heavy object with lower speed. To control the pivoting gait, a graph model predictive control is applied taking into account of these two gait modes. By adaptively switching the gait mode according to the applied external disturbance, a robot can stably perform the pivoting gait even if the external disturbance is applied to the object.

Yusuke Hayami, Weiwei Wan, Keisuke Koyama et al.

Existing methods for predicting robotic snap joint assembly cannot predict failures before their occurrence. To address this limitation, this paper proposes a method for predicting error states before the occurence of error, thereby enabling timely recovery. Robotic snap joint assembly requires precise positioning; therefore, even a slight offset between parts can lead to assembly failure. To correctly predict error states, we apply functional principal component analysis (fPCA) to 6D force/torque profiles that are terminated before the occurence of an error. The error state is identified by applying a feature vector to a decision tree, wherein the support vector machine (SVM) is employed at each node. If the estimation accuracy is low, we perform additional probing to more correctly identify the error state. Finally, after identifying the error state, a robot performs the error recovery motion based on the identified error state. Through the experimental results of assembling plastic parts with four snap joints, we show that the error states can be correctly estimated and a robot can recover from the identified error state.

Yan Wang, Cristian C. Beltran-Hernandez, Weiwei Wan et al.

Robotic assembly tasks involve complex and low-clearance insertion trajectories with varying contact forces at different stages. While the nominal motion trajectory can be easily obtained from human demonstrations through kinesthetic teaching, teleoperation, simulation, among other methods, the force profile is harder to obtain especially when a real robot is unavailable. It is difficult to obtain a realistic force profile in simulation even with physics engines. Such simulated force profiles tend to be unsuitable for the actual robotic assembly due to the reality gap and uncertainty in the assembly process. To address this problem, we present a combined learning-based framework to imitate human assembly skills through hybrid trajectory learning and force learning. The main contribution of this work is the development of a framework that combines hierarchical imitation learning, to learn the nominal motion trajectory, with a reinforcement learning-based force control scheme to learn an optimal force control policy, that can satisfy the nominal trajectory while adapting to the force requirement of the assembly task. To further improve the imitation learning part, we develop a hierarchical architecture, following the idea of goal-conditioned imitation learning, to generate the trajectory learning policy on the \textit{skill} level offline. Through experimental validations, we corroborate that the proposed learning-based framework is robust to uncertainty in the assembly task, can generate high-quality trajectories, and can find suitable force control policies, which adapt to the task's force requirements more efficiently.

Mohamed Raessa, Weiwei Wan, Kensuke Harada

This paper presents a hierarchical motion planner for planning the manipulation motion to repose long and heavy objects considering external support surfaces. The planner includes a task level layer and a motion level layer. We formulate the manipulation planning problem at the task level by considering grasp poses as nodes and object poses for edges. We consider regrasping and constrained in-hand slip (drooping) during building graphs and find mixed regrasping and drooping sequences by searching the graph. The generated sequences autonomously divide the object weight between the arm and the support surface and avoid configuration obstacles. Cartesian planning is used at the robot motion level to generate motions between adjacent critical grasp poses of the sequence found by the task level layer. Various experiments are carried out to examine the performance of the proposed planner. The results show improved capability of robot arms to manipulate long and heavy objects using the proposed planner. Our contribution is we initially develop a graph-based planning system that reasons both in-hand and regrasp manipulation motion considering external supports. On one hand, the planner integrates regrasping and drooping to realize in-hand manipulation with external support. On the other hand, it switches states by releasing and regrasping objects when the object is in stably placed. The search graphs' nodes could be retrieved from remote cloud servers that provide a large amount of pre-annotated data to implement cyber intelligence.

Shogo Hayakawa, Weiwei Wan, Keisuke Koyama et al.

This paper develops a planner that plans the action sequences and motion for a dual-arm robot to lift up and flip heavy plates using crane pulley blocks. The problem is motivated by the low payload of modern collaborative robots. Instead of directly manipulating heavy plates that collaborative robots cannot afford, the paper develops a planner for collaborative robots to operate crane pulley blocks. The planner assumes a target plate is pre-attached to the crane hook. It optimizes dual-arm action sequences and plans the robot's dual-arm motion that pulls the rope of the crane pulley blocks to lift up the plate. The crane pulley blocks reduce the payload that each robotic arm needs to bear. When the plate is lifted up to a satisfying pose, the planner plans a pushing motion for one of the robot arms to tumble over the plate while considering force and moment constraints. The article presents the technical details of the planner and several experiments and analysis carried out using a dual-arm robot made by two Universal Robots UR3 arms. The influence of various parameters and optimization goals are investigated and compared in depth. The results show that the proposed planner is flexible and efficient.

Jingren Xu, Yukiyasu Domae, Toshio Ueshiba et al.

In this paper, we present a planner that plans a sequence of base positions for a mobile manipulator to efficiently and robustly collect objects stored in distinct trays. We achieve high efficiency by exploring the common areas where a mobile manipulator can grasp objects stored in multiple trays simultaneously and move the mobile manipulator to the common areas to reduce the time needed for moving the mobile base. We ensure robustness by optimizing the base position with the best clearance to positioning uncertainty so that a mobile manipulator can complete the task even if there is a certain deviation from the planned base positions. Besides, considering different styles of object placement in the tray, we analyze feasible schemes for dynamically updating the base positions based on either the remaining objects or the target objects to be picked in one round of the tasks. In the experiment part, we examine our planner on various scenarios, including different object placement: (1) Regularly placed toy objects; (2) Randomly placed industrial parts; and different schemes for online execution: (1) Apply globally static base positions; (2) Dynamically update the base positions. The experiment results demonstrate the efficiency, robustness and feasibility of the proposed method.

Ruishuang Liu, Weiwei Wan, Keisuke Koyama et al.

This paper develops a flexible and robust robotic system for autonomous drawing on 3D surfaces. The system takes 2D drawing strokes and a 3D target surface (mesh or point clouds) as input. It maps the 2D strokes onto the 3D surface and generates a robot motion to draw the mapped strokes using visual recognition, grasp pose reasoning, and motion planning. The system is flexible compared to conventional robotic drawing systems as we do not fix drawing tools to the end of a robot arm. Instead, a robot selects drawing tools using a vision system and holds drawing tools for painting using its hand. Meanwhile, with the flexibility, the system has high robustness thanks to the following crafts: First, a high-quality mapping method is developed to minimize deformation in the strokes. Second, visual detection is used to re-estimate the drawing tool's pose before executing each drawing motion. Third, force control is employed to avoid noisy visual detection and calibration, and ensure a firm touch between the pen tip and a target surface. Fourth, error detection and recovery are implemented to deal with unexpected problems. The planning and executions are performed in a closed-loop manner until the strokes are successfully drawn. We evaluate the system and analyze the necessity of the various crafts using different real-word tasks. The results show that the proposed system is flexible and robust to generate a robot motion from picking and placing the pens to successfully drawing 3D strokes on given surfaces.

Daniel Sanchez, Weiwei Wan, Keisuke Koyama et al.

In this paper, we present a planner for a master dual-arm robot to manipulate tethered tools with an assistant dual-arm robot's help. The assistant robot provides assistance to the master robot by manipulating the tool cable and avoiding collisions. The provided assistance allows the master robot to perform tool placements on the robot workspace table to regrasp the tool, which would typically fail since the tool cable tension may change the tool positions. It also allows the master robot to perform tool handovers, which would normally cause entanglements or collisions with the cable and the environment without the assistance. Simulations and real-world experiments are performed to validate the proposed planner.

Peihao Shi, Zhengtao Hu, Kazuyuki Nagata et al.

This paper proposes an adaptive pin-array fixture. The key idea of this research is to use the shape-memorable mechanism of pin array to fix multiple different shaped parts with common pin configuration. The clamping area consists of a matrix of passively slid-able pins that conform themselves to the contour of the target object. Vertical motion of the pins enables the fixture to encase the profile of the object. The shape memorable mechanism is realized by the combination of the rubber bush and fixing mechanism of a pin. Several physical peg-in-hole tasks is conducted to verify the feasibility of the fixture.

Weiwei Wan, Takeyuki Kotaka, Kensuke Harada

The paper develops a robotic manipulation system to treat the pressing needs for handling a large number of test tubes in clinical examination and replace or reduce human labor. It presents the technical details of the system, which separates and arranges test tubes in racks with the help of 3D vision and artificial intelligence (AI) reasoning/planning. The developed system only requires a person to put a rack with mixed and non-arranged tubes in front of a robot. The robot autonomously performs recognition, reasoning, planning, manipulation, etc., and returns a rack with separated and arranged tubes. The system is simple-to-use, and there are no requests for expert knowledge in robotics. We expect such a system to play an important role in helping managing public health and hope similar systems could be extended to other clinical manipulation like handling mixers and pipettes in the future.

Weibo Huang, Hong Liu, Weiwei Wan

This paper presents an online initialization method for bootstrapping the optimization-based monocular visual-inertial odometry (VIO). The method can online calibrate the relative transformation (spatial) and time offsets (temporal) among camera and IMU, as well as estimate the initial values of metric scale, velocity, gravity, gyroscope bias, and accelerometer bias during the initialization stage. To compensate for the impact of time offset, our method includes two short-term motion interpolation algorithms for the camera and IMU pose estimation. Besides, it includes a three-step process to incrementally estimate the parameters from coarse to fine. First, the extrinsic rotation, gyroscope bias, and time offset are estimated by minimizing the rotation difference between the camera and IMU. Second, the metric scale, gravity, and extrinsic translation are approximately estimated by using the compensated camera poses and ignoring the accelerometer bias. Third, these values are refined by taking into account the accelerometer bias and the gravitational magnitude. For further optimizing the system states, a nonlinear optimization algorithm, which considers the time offset, is introduced for global and local optimization. Experimental results on public datasets show that the initial values and the extrinsic parameters, as well as the sensor poses, can be accurately estimated by the proposed method.

Kazuki Higashi, Keisuke Koyama, Ryuta Ozawa et al.

Synergy supplies a practical approach for expressing various postures of a multi-fingered hand. However, a conventional synergy defined for reproducing grasping postures cannot perform general-purpose tasks expected for a multi-fingered hand. Locking the position of particular fingers is essential for a multi-fingered hand to manipulate an object. When using conventional synergy based control to manipulate an object, which requires locking some fingers, the coordination of joints is heavily restricted, decreasing the dexterity of the hand. We propose a functionally divided manipulation synergy (FDMS) method, which provides a synergy-based control to achieves both dimensionality reduction and in-hand manipulation. In FDMS, first, we define the function of each finger of the hand as either "manipulation" or "fixed." Then, we apply synergy control only to the fingers having the manipulation function, so that dexterous manipulations can be realized with few control inputs. The effectiveness of our proposed approach is experimentally verified.

Jingren Xu, Weiwei Wan, Keisuke Koyama et al.

In this paper, we present a structured approach to selecting and designing a set of grippers for an assembly task. Compared to current experience-based gripper design method, our approach accelerates the design process by automatically generating a set of initial design options on gripper type and parameters according to the CAD models of assembly components. We use mesh segmentation techniques to segment the assembly components and fit the segmented parts with shape primitives, according to the predefined correspondence between primitive shape and gripper type, suitable gripper types and parameters can be selected and extracted from the fitted shape primitives. Moreover, we incorporate the assembly constraints in the further evaluation of the initially obtained gripper types and parameters. Considering the affordance of the segmented parts and the collision avoidance between the gripper and the subassemblies, applicable gripper types and parameters can be filtered from the initial options. Among the applicable gripper configurations, we further optimize number of grippers for performing the assembly task, by exploring the gripper that is able to handle multiple assembly components during the assembly. Finally, the feasibility of the designed grippers is experimentally verified by assembling a part of an industrial product.

Hao Chen, Weiwei Wan, Keisuke Koyama et al.

This paper presents a planner that can automatically find an optimal assembly sequence for a dual-arm robot to assemble the soma blocks. The planner uses the mesh model of objects and the final state of the assembly to generate all possible assembly sequence and evaluate the optimal assembly sequence by considering the stability, graspability, assemblability, as well as the need for a second arm. Especially, the need for a second arm is considered when supports from worktables and other workpieces are not enough to produce a stable assembly. The planner will refer to an assisting grasp to additionally hold and support the unstable components so that the robot can further assemble new workpieces and finally reach a stable state. The output of the planner is the optimal assembly orders, candidate grasps, assembly directions, and the assisting grasps if any. The output of the planner can be used to guide a dual-arm robot to perform the assembly task. The planner is verified in both simulations and real-world executions.

Jingren Xu, Kensuke Harada, Weiwei Wan et al.

In this paper, we address efficiently and robustly collecting objects stored in different trays using a mobile manipulator. A resolution complete method, based on precomputed reachability database, is proposed to explore collision-free inverse kinematics (IK) solutions and then a resolution complete set of feasible base positions can be determined. This method approximates a set of representative IK solutions that are especially helpful when solving IK and checking collision are treated separately. For real world applications, we take into account the base positioning uncertainty and plan a sequence of base positions that reduce the number of necessary base movements for collecting the target objects, the base sequence is robust in that the mobile manipulator is able to complete the part-supply task even there is certain deviation from the planned base positions. Our experiments demonstrate both the efficiency compared to regular base sequence and the feasibility in real world applications.

Felix von Drigalski, Christian Schlette, Martin Rudorfer et al.

The Industrial Assembly Challenge at the World Robot Summit was held in 2018 to showcase the state-of-the-art of autonomous manufacturing systems. The challenge included various tasks, such as bin picking, kitting, and assembly of standard industrial parts into 2D and 3D assemblies. Some of the tasks were only revealed at the competition itself, representing the challenge of "level 5" automation, i. e., programming and setting up an autonomous assembly system in less than one day. We conducted a survey among the teams that participated in the challenge and investigated aspects such as team composition, development costs, system setups as well as the teams' strategies and approaches. An analysis of the survey results reveals that the competitors have been in two camps: those constructing conventional robotic work cells with off-the-shelf tools, and teams who mostly relied on custom-made end effectors and novel software approaches in combination with collaborative robots. While both camps performed reasonably well, the winning team chose a middle ground in between, combining the efficiency of established play-back programming with the autonomy gained by CAD-based object detection and force control for assembly operations.

Yan Wang, Kensuke Harada, Weiwei Wan

Complex and skillful motions in actual assembly process are challenging for the robot to generate with existing motion planning approaches, because some key poses during the human assembly can be too skillful for the robot to realize automatically. In order to deal with this problem, this paper develops a motion planning method using skillful motions from demonstration, which can be applied to complete robotic assembly process including complex and skillful motions. In order to demonstrate conveniently without redundant third-party devices, we attach augmented reality (AR) markers to the manipulated object to track and capture poses of the object during the human assembly process, which are employed as key poses to execute motion planning by the planner. Derivative of every key pose serves as criterion to determine the priority of use of key poses in order to accelerate the motion planning. The effectiveness of the presented method is verified through some numerical examples and actual robot experiments.

Mohamed Raessa, Jimmy Chi Yin Chen, Weiwei Wan et al.

This paper develops a robotic manipulation planner for human-robot collaborative assembly. Unlike previous methods which study an independent and fully AI-equipped autonomous system, this paper explores the subtask distribution between a robot and a human and studies a human-in-the-loop robotic system for collaborative assembly. The system distributes the subtasks of an assembly to robots and humans by exploiting their advantages and avoiding their disadvantages. The robot in the system will work on pick-and-place tasks and provide workpieces to humans. The human collaborator will work on fine operations like aligning, fixing, screwing, etc. A constraint based incremental manipulation planning method is proposed to generate the motion for the robots. The performance of the proposed system is demonstrated by asking a human and the dual-arm robot to collaboratively assemble a cabinet. The results showed that the proposed system and planner are effective, efficient, and can assist humans in finishing the assembly task comfortably.

Daniel Sanchez, Weiwei Wan, Kensuke Harada

In this paper, we present a planner for manipulating tethered tools using dual-armed robots. The planner generates robot motion sequences to maneuver a tool and its cable while avoiding robot-cable entanglements. Firstly, the planner generates an Object Manipulation Motion Sequence (OMMS) to handle the tool and place it in desired poses. Secondly, the planner examines the tool movement associated with the OMMS and computes candidate positions for a cable slider, to maneuver the tool cable and avoid collisions. Finally, the planner determines the optimal slider positions to avoid entanglements and generates a Cable Manipulation Motion Sequence (CMMS) to place the slider in these positions. The robot executes both the OMMS and CMMS to handle the tool and its cable to avoid entanglements and excess cable bending. Simulations and real-world experiments help validate the proposed method.

Hao Chen, Weiwei Wan, Kensuke Harada

This paper proposes a combined task and motion planner for a dual-arm robot to use a suction cup tool. The planner consists of three sub-planners -- A suction pose sub-planner and two regrasp and motion sub-planners. The suction pose sub-planner finds all the available poses for a suction cup tool to suck on the object, using the models of the tool and the object. The regrasp and motion sub-planner builds the regrasp graph that represents all possible grasp sequences to reorient and move the suction cup tool from an initial pose to a goal pose. Two regrasp graphs are used to plan for a single suction cup and the complex of the suction cup and an object respectively. The output of the proposed planner is a sequence of robot motion that uses a suction cup tool to manipulate objects following human instructions. The planner is examined and analyzed by both simulation experiments and real-world executions using several real-world tasks. The results show that the planner is efficient, robust, and can generate sequential transit and transfer robot motion to finish complicated combined task and motion planning tasks in a few seconds.

Weiwei Wan, Kensuke Harada, Fumio Kanehiro

This paper develops model-based grasp planning algorithms for assembly tasks. It focuses on industrial end-effectors like grippers and suction cups, and plans grasp configurations considering CAD models of target objects. The developed algorithms are able to stably plan a large number of high-quality grasps, with high precision and little dependency on the quality of CAD models. The undergoing core technique is superimposed segmentation, which pre-processes a mesh model by peeling it into facets. The algorithms use superimposed segments to locate contact points and parallel facets, and synthesize grasp poses for popular industrial end-effectors. Several tunable parameters were prepared to adapt the algorithms to meet various requirements. The experimental section demonstrates the advantages of the algorithms by analyzing the cost and stability of the algorithms, the precision of the planned grasps, and the tunable parameters with both simulations and real-world experiments. Also, some examples of robotic assembly systems using the proposed algorithms are presented to demonstrate the efficacy.

Ryota Moriyama, Weiwei Wan, Kensuke Harada

Planning dual-arm assembly of more than three objects is a challenging Task and Motion Planning (TAMP) problem. The assembly planner shall consider not only the pose constraints of objects and robots, but also the gravitational constraints that may break the finished part. This paper proposes a planner to plan the dual-arm assembly of more than three objects. It automatically generates the grasp configurations and assembly poses, and simultaneously searches and backtracks the grasp space and assembly space to accelerate the motion planning of robot arms. Meanwhile, the proposed method considers gravitational constraints during robot motion planning to avoid breaking the finished part. In the experiments and analysis section, the time cost of each process and the influence of different parameters used in the proposed planner are compared and analyzed. The optimal values are used to perform real-world executions of various robotic assembly tasks. The planner is proved to be robust and efficient through the experiments.

Joshua C. Triyonoputro, Weiwei Wan, Kensuke Harada

This paper uses robots to assemble pegs into holes on surfaces with different colors and textures. It especially targets at the problem of peg-in-hole assembly with initial position uncertainty. Two in-hand cameras and a force-torque sensor are used to account for the position uncertainty. A program sequence comprising learning-based visual servoing, spiral search, and impedance control is implemented to perform the peg-in-hole task with feedback from the above sensors. Contributions are mainly made in the learning-based visual servoing of the sequence, where a deep neural network is trained with various sets of synthetic data generated using the concept of domain randomization to predict where a hole is. In the experiments and analysis section, the network is analyzed and compared, and a real-world robotic system to insert pegs to holes using the proposed method is implemented. The results show that the implemented peg-in-hole assembly system can perform successful peg-in-hole insertions on surfaces with various colors and textures. It can generally speed up the entire peg-in-hole process.

Zhengtao Hu, Weiwei Wan, Kensuke Harada

This work designs a mechanical tool for robots with 2-finger parallel grippers, which extends the function of the robotic gripper without additional requirements on tool exchangers or other actuators. The fundamental kinematic structure of the mechanical tool is two symmetric parallelograms which transmit the motion of the robotic gripper to the mechanical tool. Four torsion springs are attached to the four inner joints of the two parallelograms to open the tool as the robotic gripper releases. The forces and transmission are analyzed in detail to make sure the tool reacts well with respect to the gripping forces and the spring stiffness. Also, based on the kinematic structure, variety tooltips were designed for the mechanical tool to perform various tasks. The kinematic structure can be a platform to apply various skillful gripper designs. The designed tool could be treated as a normal object and be picked up and used by automatically planned grasps. A robot may locate the tool through the AR markers attached to the tool body, grasp the tool by selecting an automatically planned grasp, and move the tool from any arbitrary pose to a specific pose to grip objects. The robot may also determine the optimal grasps and usage according to the requirements of given tasks.

Daniel Sanchez, Weiwei Wan, Kensuke Harada

Robotic manipulation of tethered tools is widely seen in robotic work cells. They may cause excess strain on the tool's cable or undesired entanglements with the robot's arms. This paper presents a manipulation planner with cable orientation constraints for tethered tools suspended by tool balancers. The planner uses orientation constraints to limit the bending of the balancer's cable while the robot manipulates a tool and places it in a desired pose. The constraints reduce entanglements and decrease the torque induced by the cable on the robot joints. Simulation and real-world experiments show that the constrained planner can successfully plan robot motions for the manipulation of suspended tethered tools preventing the robot from damaging the cable or getting its arms entangled, potentially avoiding accidents. The planner is expected to play promising roles in manufacturing cells.

Weiwei Wan, Kensuke Harada, Fumio Kanehiro

This paper presents a manipulation planning algorithm for robots to reorient objects. It automatically finds a sequence of robot motion that manipulates and prepares an object for specific tasks. Examples of the preparatory manipulation planning problems include reorienting an electric drill to cut holes, reorienting workpieces for assembly, and reorienting cargo for packing, etc. The proposed algorithm could plan single and dual arm manipulation sequences to solve the problems. The mechanism under the planner is a regrasp graph which encodes grasp configurations and object poses. The algorithms search the graph to find a sequence of robot motion to reorient objects. The planner is able to plan both single and dual arm manipulation. It could also automatically determine whether to use a single arm, dual arms, or their combinations to finish given tasks. The planner is examined by various humanoid robots like Nextage, HRP2Kai, HRP5P, etc., using both simulation and real-world experiments.

Hitoe Ochi, Weiwei Wan, Yajue Yang et al.

We present bilateral teleoperation system for task learning and robot motion generation. Our system includes a bilateral teleoperation platform and a deep learning software. The deep learning software refers to human demonstration using the bilateral teleoperation platform to collect visual images and robotic encoder values. It leverages the datasets of images and robotic encoder information to learn about the inter-modal correspondence between visual images and robot motion. In detail, the deep learning software uses a combination of Deep Convolutional Auto-Encoders (DCAE) over image regions, and Recurrent Neural Network with Long Short-Term Memory units (LSTM-RNN) over robot motor angles, to learn motion taught be human teleoperation. The learnt models are used to predict new motion trajectories for similar tasks. Experimental results show that our system has the adaptivity to generate motion for similar scooping tasks. Detailed analysis is performed based on failure cases of the experimental results. Some insights about the cans and cannots of the system are summarized.

Daniel Sanchez, Weiwei Wan, Kensuke Harada et al.

This paper presents a Center of Mass (CoM) based manipulation and regrasp planner that implements stability constraints to preserve the robot balance. The planner provides a graph of IK-feasible, collision-free and stable motion sequences, constructed using an energy based motion planning algorithm. It assures that the assembly motions are stable and prevent the robot from falling while performing dexterous tasks in different situations. Furthermore, the constraints are also used to perform an RRT-inspired task-related stability estimation in several simulations. The estimation can be used to select between single-arm and dual-arm regrasping configurations to achieve more stability and robustness for a given manipulation task. To validate the planner and the task-related stability estimations, several tests are performed in simulations and real-world experiments involving the HRP5P humanoid robot, the 5th generation of the HRP robot family. The experiment results suggest that the planner and the task-related stability estimation provide robust behavior for the humanoid robot while performing regrasp tasks.

Kaidi Nie, Weiwei Wan, Kensuke Harada

This paper proposes a novel robotic hand design for assembly tasks. The idea is to combine two simple grippers -- an inner gripper which is used for precise alignment, and an outer gripper which is used for stable holding. Conventional robotic hands require complicated compliant mechanisms or complicated control strategy and force sensing to conduct assemble tasks, which makes them costly and difficult to pick and arrange small objects like screws or washers. Compared to the conventional hands, the proposed design provides a low-cost solution for aligning, picking up, and arranging various objects by taking advantages of the geometric constraints of the positioning fingers and gravity. It is able to deal with small screws and washers, and eliminate the position errors of cylindrical objects or objects with cylindrical holes. In the experiments, both real-world tasks and quantitative analysis are performed to validate the aligning, picking, and arrangements abilities of the design.

Joshua C. Triyonoputro, Weiwei Wan, Kensuke Harada

This paper presents a double jaw hand for industrial assembly. The hand comprises two orthogonal parallel grippers with different mechanisms. The inner gripper is made of a crank-slider mechanism which is compact and able to firmly hold objects like shafts. The outer gripper is made of a parallelogram that has large stroke to hold big objects like pulleys. The two grippers are connected by a prismatic joint along the hand's approaching vector. The hand is able to hold two objects and perform in-hand manipulation like pull-in (insertion) and push-out (ejection). This paper presents the detailed design and implementation of the hand, and demonstrates the advantages by performing experiments on two sets of peg-in-multi-hole assembly tasks as parts of the World Robot Challenge (WRC) 2018 using a bimanual robot.

Kensuke Harada, Kento Nakayama, Weiwei Wan et al.

This paper proposes a novel assembly planner for a manipulator which can simultaneously plan assembly sequence, robot motion, grasping configuration, and exchange of grippers. Our assembly planner assumes multiple grippers and can automatically selects a feasible one to assemble a part. For a given AND/OR graph of an assembly task, we consider generating the assembly graph from which assembly motion of a robot can be planned. The edges of the assembly graph are composed of three kinds of paths, i.e., transfer/assembly paths, transit paths and tool exchange paths. In this paper, we first explain the proposed method for planning assembly motion sequence including the function of gripper exchange. Finally, the effectiveness of the proposed method is confirmed through some numerical examples and a physical experiment.

Ryo Matsumura, Kensuke Harada, Yukiyasu Domae et al.

In this research, we tackle the problem of picking an object from randomly stacked pile. Since complex physical phenomena of contact among objects and fingers makes it difficult to perform the bin-picking with high success rate, we consider introducing a learning based approach. For the purpose of collecting enough number of training data within a reasonable period of time, we introduce a physics simulator where approximation is used for collision checking. In this paper, we first formulate the learning based robotic bin-picking by using CNN (Convolutional Neural Network). We also obtain the optimum grasping posture of parallel jaw gripper by using CNN. Finally, we show that the effect of approximation introduced in collision checking is relaxed if we use exact 3D model to generate the depth image of the pile as an input to CNN.

Kensuke Harada, Weiwei Wan, Tokuo Tsuji et al.

This paper shows experimental results on learning based randomized bin-picking combined with iterative visual recognition. We use the random forest to predict whether or not a robot will successfully pick an object for given depth images of the pile taking the collision between a finger and a neighboring object into account. For the discriminator to be accurate, we consider estimating objects' poses by merging multiple depth images of the pile captured from different points of view by using a depth sensor attached at the wrist. We show that, even if a robot is predicted to fail in picking an object with a single depth image due to its large occluded area, it is finally predicted as success after merging multiple depth images. In addition, we show that the random forest can be trained with the small number of training data.

Chaoqun Wang, Lili Meng, Sizhen She et al.

Robots are increasingly operating in indoor environments designed for and shared with people. However, robots working safely and autonomously in uneven and unstructured environments still face great challenges. Many modern indoor environments are designed with wheelchair accessibility in mind. This presents an opportunity for wheeled robots to navigate through sloped areas while avoiding staircases. In this paper, we present an integrated software and hardware system for autonomous mobile robot navigation in uneven and unstructured indoor environments. This modular and reusable software framework incorporates capabilities of perception and navigation. Our robot first builds a 3D OctoMap representation for the uneven environment with the 3D mapping using wheel odometry, 2D laser and RGB-D data. Then we project multilayer 2D occupancy maps from OctoMap to generate the the traversable map based on layer differences. The safe traversable map serves as the input for efficient autonomous navigation. Furthermore, we employ a variable step size Rapidly Exploring Random Trees that could adjust the step size automatically, eliminating tuning step sizes according to environments. We conduct extensive experiments in simulation and real-world, demonstrating the efficacy and efficiency of our system.

Weiwei Wan, Kensuke Harada

This paper develops intelligent algorithms for robots to reorient objects. Given the initial and goal poses of an object, the proposed algorithms plan a sequence of robot poses and grasp configurations that reorient the object from its initial pose to the goal. While the topic has been studied extensively in previous work, this paper makes important improvements in grasp planning by using over-segmented meshes, in data storage by using relational database, and in regrasp planning by mixing real-world roadmaps. The improvements enable robots to do robust regrasp planning using 10,000s of grasps and their relationships in interactive time. The proposed algorithms are validated using various objects and robots.

Weiwei Wan, Kensuke Harada, Kazuyuki Nagata

This paper develops a planner to find an optimal assembly sequence to assemble several objects. The input to the planner is the mesh models of the objects, the relative poses between the objects in the assembly, and the final pose of the assembly. The output is an optimal assembly sequence, namely (1) in which order should one assemble the objects, (2) from which directions should the objects be dropped, and (3) candidate grasps of each object. The proposed planner finds the optimal solution by automatically permuting, evaluating, and searching the possible assembly sequences considering stability, graspability, and assemblability qualities. It is expected to guide robots to do assembly using translational motion. The output provides initial and goal configurations to motion planning algorithms. It is ready to be used by robots and is demonstrated using several simulations and real-world executions.

Weiwei Wan, Hisashi Igawa, Kensuke Harada et al.

This paper presents a mid-level planning system for object reorientation. It includes a grasp planner, a placement planner, and a regrasp sequence solver. Given the initial and goal poses of an object, the mid-level planning system finds a sequence of hand configurations that reorient the object from the initial to the goal. This mid-level planning system is open to low-level motion planning algorithm by providing two end-effector poses as the input. It is also open to high-level symbolic planners by providing interface functions like placing an object to a given position at a given rotation. The planning system is demonstrated with several simulation examples and real-robot executions using a Kawada Hiro robot and Robotiq 85 grippers.

Kensuke Harada, Weiwei Wan, Tokuo Tsuji et al.

This paper proposes a iterative visual recognition system for learning based randomized bin-picking. Since the configuration on randomly stacked objects while executing the current picking trial is just partially different from the configuration while executing the previous picking trial, we consider detecting the poses of objects just by using a part of visual image taken at the current picking trial where it is different from the visual image taken at the previous picking trial. By using this method, we do not need to try to detect the poses of all objects included in the pile at every picking trial. Assuming the 3D vision sensor attached at the wrist of a manipulator, we first explain a method to determine the pose of a 3D vision sensor maximizing the visibility of randomly stacked objects. Then, we explain a method for detecting the poses of randomly stacked objects. Effectiveness of our proposed approach is confirmed by experiments using a dual-arm manipulator where a 3D vision sensor and the two-fingered hand attached at the right and the left wrists, respectively.