Toshiaki Tsuji

Toshiaki Tsuji, Yasuhiro Kato

In contact-rich tasks, setting the stiffness of the control system is a critical factor in its performance. Although the setting range can be extended by making the stiffness matrix asymmetric, its stability has not been proven. This study focuses on the stability of compliance control in a robot arm that deals with an asymmetric stiffness matrix. It discusses the convergence stability of the admittance control. The paper explains how to derive an asymmetric stiffness matrix and how to incorporate it into the admittance model. The authors also present simulation and experimental results that demonstrate the effectiveness of their proposed method.

Toshiaki Tsuji, Yasuhiro Kato, Gokhan Solak et al.

This paper comprehensively surveys research trends in imitation learning for contact-rich robotic tasks. Contact-rich tasks, which require complex physical interactions with the environment, represent a central challenge in robotics due to their nonlinear dynamics and sensitivity to small positional deviations. The paper examines demonstration collection methodologies, including teaching methods and sensory modalities crucial for capturing subtle interaction dynamics. We then analyze imitation learning approaches, highlighting their applications to contact-rich manipulation. Recent advances in multimodal learning and foundation models have significantly enhanced performance in complex contact tasks across industrial, household, and healthcare domains. Through systematic organization of current research and identification of challenges, this survey provides a foundation for future advancements in contact-rich robotic manipulation.

Koki Yamane, Yunhan Li, Masashi Konosu et al.

In recent years, the advancement of imitation learning has led to increased interest in teleoperating low-cost manipulators to collect demonstration data. However, most existing systems rely on unilateral control, which only transmits target position values. While this approach is easy to implement and suitable for slow, non-contact tasks, it struggles with fast or contact-rich operations due to the absence of force feedback. This work demonstrates that fast teleoperation with force feedback is feasible even with force-sensorless, low-cost manipulators by leveraging 4-channel bilateral control. Based on accurately identified manipulator dynamics, our method integrates nonlinear terms compensation, velocity and external force estimation, and variable gain corresponding to inertial variation. Furthermore, using data collected by 4-channel bilateral control, we show that incorporating force information into both the input and output of learned policies improves performance in imitation learning. These results highlight the practical effectiveness of our system for high-fidelity teleoperation and data collection on affordable hardware.

Nozomu Masuya, Hiroshi Sato, Koki Yamane et al.

Because imitation learning relies on human demonstrations in hard-to-simulate settings, the inclusion of force control in this method has resulted in a shortage of training data, even with a simple change in speed. Although the field of data augmentation has addressed the lack of data, conventional methods of data augmentation for robot manipulation are limited to simulation-based methods or downsampling for position control. This paper proposes a novel method of data augmentation that is applicable to force control and preserves the advantages of real-world datasets. We applied teaching-playback at variable speeds as real-world data augmentation to increase both the quantity and quality of environmental reactions at variable speeds. An experiment was conducted on bilateral control-based imitation learning using a method of imitation learning equipped with position-force control. We evaluated the effect of real-world data augmentation on two tasks, pick-and-place and wiping, at variable speeds, each from two human demonstrations at fixed speed. The results showed a maximum 55% increase in success rate from a simple change in speed of real-world reactions and improved accuracy along the duration/frequency command by gathering environmental reactions at variable speeds.

Hiroshi Sato, Masashi Konosu, Sho Sakaino et al.

In recent years, imitation learning using neural networks has enabled robots to perform flexible tasks. However, since neural networks operate in a feedforward structure, they do not possess a mechanism to compensate for output errors. To address this limitation, we developed a feedback mechanism to correct these errors. By employing a hierarchical structure for neural networks comprising lower and upper layers, the lower layer was controlled to follow the upper layer. Additionally, using a multi-layer perceptron in the lower layer, which lacks an internal state, enhanced the error feedback. In the character-writing task, this model demonstrated improved accuracy in writing previously untrained characters. In the character-writing task, this model demonstrated improved accuracy in writing previously untrained characters. Through autonomous control with error feedback, we confirmed that the lower layer could effectively track the output of the upper layer. This study represents a promising step toward integrating neural networks with control theories.

Koki Yamane, Sho Sakaino, Toshiaki Tsuji

Imitation learning enables robots to learn and replicate human behavior from training data. Recent advances in machine learning enable end-to-end learning approaches that directly process high-dimensional observation data, such as images. However, these approaches face a critical challenge when processing data from multiple modalities, inadvertently ignoring data with a lower correlation to the desired output, especially when using short sampling periods. This paper presents a useful method to address this challenge, which amplifies the influence of data with a relatively low correlation to the output by inputting the data into each neural network layer. The proposed approach effectively incorporates diverse data sources into the learning process. Through experiments using a simple pick-and-place operation with raw images and joint information as input, significant improvements in success rates are demonstrated even when dealing with data from short sampling periods.

Masahiro Aita, Keito Sugawara, Sho Sakaino et al.

In this study, we propose a method to model the local and global features of the drawing/grinding trajectory with hierarchical Variational Autoencoders (VAEs). By combining two separately trained VAE models in a hierarchical structure, it is possible to generate trajectories with high reproducibility for both local and global features. The hierarchical generation network enables the generation of higher-order trajectories with a relatively small amount of training data. The simulation and experimental results demonstrate the generalization performance of the proposed method. In addition, we confirmed that it is possible to generate new trajectories, which have never been learned in the past, by changing the combination of the learned models.

Kazuki Hayashi, Ayumu Sasagawa, Sho Sakaino et al.

In the near future, robots are expected to work with humans or operate alone and may replace human workers in various fields such as homes and factories. In a previous study, we proposed bilateral control-based imitation learning that enables robots to utilize force information and operate almost simultaneously with an expert's demonstration. In addition, we recently proposed an autoregressive neural network model (SM2SM) for bilateral control-based imitation learning to obtain long-term inferences. In the SM2SM model, both master and slave states must be input, but the master states are obtained from the previous outputs of the SM2SM model, resulting in destabilized estimation under large environmental variations. Hence, a new autoregressive neural network model (S2SM) is proposed in this study. This model requires only the slave state as input and its outputs are the next slave and master states, thereby improving the task success rates. In addition, a new feedback controller that utilizes the error between the responses and estimates of the slave is proposed, which shows better reproducibility.

Yuki Saigusa, Ayumu Sasagawa, Sho Sakaino et al.

Robotic motion generation methods using machine learning have been studied in recent years. Bilateral control-based imitation learning can imitate human motions using force information. By means of this method, variable speed motion generation that considers physical phenomena such as the inertial force and friction can be achieved. However, the previous study only focused on a simple reciprocating motion. To learn the complex relationship between the force and speed more accurately, it is necessary to learn multiple actions using many joints. In this paper, we propose a variable speed motion generation method for multiple motions. We considered four types of neural network models for the motion generation and determined the best model for multiple motions at variable speeds. Subsequently, we used the best model to evaluate the reproducibility of the task completion time for the input completion time command. The results revealed that the proposed method could change the task completion time according to the specified completion time command in multiple motions.

Sho Sakaino, Kazuki Fujimoto, Yuki Saigusa et al.

The generation of robot motions in the real world is difficult by using conventional controllers alone and requires highly intelligent processing. In this regard, learning-based motion generations are currently being investigated. However, the main issue has been improvements of the adaptability to spatially varying environments, but a variation of the operating speed has not been investigated in detail. In contact-rich tasks, it is especially important to be able to adjust the operating speed because a nonlinear relationship occurs between the operating speed and force (e.g., inertial and frictional forces), and it affects the results of the tasks. Therefore, in this study, we propose a method for generating variable operating speeds while adapting to spatial perturbations in the environment. The proposed method can be adapted to nonlinearities by utilizing a small amount of motion data. We experimentally evaluated the proposed method by erasing a line using an eraser fixed to the tip of the robot as an example of a contact-rich task. Furthermore, the proposed method enables a robot to perform a task faster than a human operator and is capable of operating close to the control bandwidth.

Ayumu Sasagawa, Sho Sakaino, Toshiaki Tsuji

Robots that can execute various tasks automatically on behalf of humans are becoming an increasingly important focus of research in the field of robotics. Imitation learning has been studied as an efficient and high-performance method, and imitation learning based on bilateral control has been proposed as a method that can realize fast motion. However, because this method cannot implement autoregressive learning, this method may not generate desirable long-term behavior. Therefore, in this paper, we propose a method of autoregressive learning for bilateral control-based imitation learning. A new neural network model for implementing autoregressive learning is proposed. In this study, three types of experiments are conducted to verify the effectiveness of the proposed method. The performance is improved compared to conventional approaches; the proposed method has the highest rate of success. Owing to the structure and autoregressive learning of the proposed model, the proposed method can generate the desirable motion for successful tasks and have a high generalization ability for environmental changes.

Masahide Oikawa, Kyo Kutsuzawa, Sho Sakaino et al.

There is an increased demand for task automation in robots. Contact-rich tasks, wherein multiple contact transitions occur in a series of operations, are extensively being studied to realize high accuracy. In this study, we propose a methodology that uses reinforcement learning (RL) to achieve high performance in robots for the execution of assembly tasks that require precise contact with objects without causing damage. The proposed method ensures the online generation of stiffness matrices that help improve the performance of local trajectory optimization. The method has an advantage of rapid response owing to short sampling time of the trajectory planning. The effectiveness of the method was verified via experiments involving two contact-rich tasks. The results indicate that the proposed method can be implemented in various contact-rich manipulations. A demonstration video shows the performance. (https://youtu.be/gxSCl7Tp4-0)

Ayumu Sasagawa, Kazuki Fujimoto, Sho Sakaino et al.

Robots are required to autonomously respond to changing situations. Imitation learning is a promising candidate for achieving generalization performance, and extensive results have been demonstrated in object manipulation. However, cooperative work between humans and robots is still a challenging issue because robots must control dynamic interactions among themselves, humans, and objects. Furthermore, it is difficult to follow subtle perturbations that may occur among coworkers. In this study, we find that cooperative work can be accomplished by imitation learning using bilateral control. Thanks to bilateral control, which can extract response values and command values independently, human skills to control dynamic interactions can be extracted. Then, the task of serving food is considered. The experimental results clearly demonstrate the importance of force control, and the dynamic interactions can be controlled by the inferred action force.

Kazuki Fujimoto, Sho Sakaino, Toshiaki Tsuji

Various adaptive abilities are required for robots interacting with humans in daily life. It is difficult to design adaptive algorithms manually; however, by using end-to-end machine learning, labor can be saved during the design process. In our previous research, a task requiring force adjustment was achieved through imitation learning that considered position and force information using a four-channel bilateral control. Unfortunately, tasks that include long-term (slow) motion are still challenging. Furthermore, during system identification, there is a method known as the multi-decimation (MD) identification method. It separates lower and higher frequencies, and then identifies the parameters characterized at each frequency. Therefore, we proposed utilizing machine learning to take advantage of the MD method to infer short-term and long-term (high and low frequency, respectively) motion. In this paper, long-term motion tasks such as writing a letter using a pen fixed on a robot are discussed. We found differences in suitable sampling periods between position and force information. The validity of the proposed method was then experimentally verified, showing the importance of long-term inference with adequate sampling periods.

Tsuyoshi Adachi, Kazuki Fujimoto, Sho Sakaino et al.

This study proposes an imitation learning method based on force and position information. Force information is required for precise object manipulation but is difficult to obtain because the acting and reaction forces cannnot be separated. To separate the forces, we proposed to introduce bilateral control, in which the acting and reaction forces are divided using two robots. In the proposed method, two models of neural networks learn a task; to draw a line along a ruler. We verify the possibility that force information is essential to imitate the human skill of object manipulation.

Tomoya Kitamura, Yuu Hasegawa, Sho Sakaino et al.

In this study, we developed a method to estimate the relationship between stimulation current and volatility during isometric contraction. In functional electrical stimulation (FES), joints are driven by applying voltage to muscles. This technology has been used for a long time in the field of rehabilitation, and recently application oriented research has been reported. However, estimation of the relationship between stimulus value and exercise capacity has not been discussed to a great extent. Therefore, in this study, a human muscle model was estimated using the transfer function estimation method with fast Fourier transform. It was found that the relationship between stimulation current and force exerted could be expressed by a first-order lag system. In verification of the force estimate, the ability of the proposed model to estimate the exerted force under steady state response was found to be good.