Nima Fazeli

Salma Mozaffari, Daniel Ruan, William van den Bogert et al. · princeton

Fabrication uncertainty arising from tolerance accumulation, material imperfection, and positioning errors remains a critical barrier to automated robotic assembly in construction, particularly for contact-rich manipulation tasks governed by friction and geometric constraints. This paper investigates the deployment of diffusion policy learning on construction-scale industrial robots to enable robust, high-precision assembly under such uncertainty, using tight-fitting mortise and tenon timber joinery as a representative case study. Sensory-motor diffusion policies are trained using teleoperated demonstrations collected from an industrial robotic workcell equipped with force/torque sensing. A two-phase experimental study evaluates baseline performance and robustness under randomized positional perturbations up to 10 mm, far exceeding the sub-millimeter joint clearance. The best-performing policy achieved 100% success under nominal conditions and 75% average success under uncertainty. These results provide initial evidence that diffusion policies compensate for misalignments through contact-aware control, representing a step toward robust robotic assembly in construction under tight tolerances.

Yizhou Chen, Andrea Sipos, Mark Van der Merwe et al.

Learning representations in the joint domain of vision and touch can improve manipulation dexterity, robustness, and sample-complexity by exploiting mutual information and complementary cues. Here, we present Visuo-Tactile Transformers (VTTs), a novel multimodal representation learning approach suited for model-based reinforcement learning and planning. Our approach extends the Visual Transformer \cite{dosovitskiy2021image} to handle visuo-tactile feedback. Specifically, VTT uses tactile feedback together with self and cross-modal attention to build latent heatmap representations that focus attention on important task features in the visual domain. We demonstrate the efficacy of VTT for representation learning with a comparative evaluation against baselines on four simulated robot tasks and one real world block pushing task. We conduct an ablation study over the components of VTT to highlight the importance of cross-modality in representation learning.

Yinpei Dai, Jayjun Lee, Nima Fazeli et al.

Developing robust and correctable visuomotor policies for robotic manipulation is challenging due to the lack of self-recovery mechanisms from failures and the limitations of simple language instructions in guiding robot actions. To address these issues, we propose a scalable data generation pipeline that automatically augments expert demonstrations with failure recovery trajectories and fine-grained language annotations for training. We then introduce Rich languAge-guided failure reCovERy (RACER), a supervisor-actor framework, which combines failure recovery data with rich language descriptions to enhance robot control. RACER features a vision-language model (VLM) that acts as an online supervisor, providing detailed language guidance for error correction and task execution, and a language-conditioned visuomotor policy as an actor to predict the next actions. Our experimental results show that RACER outperforms the state-of-the-art Robotic View Transformer (RVT) on RLbench across various evaluation settings, including standard long-horizon tasks, dynamic goal-change tasks and zero-shot unseen tasks, achieving superior performance in both simulated and real world environments. Videos and code are available at: https://rich-language-failure-recovery.github.io.

William van den Bogert, Madhavan Iyengar, Nima Fazeli

Tactile sensing is a widely-studied means of implicit communication between robot and human. In this paper, we investigate how tactile sensing can help bridge differences between robotic embodiments in the context of collaborative manipulation. For a robot, learning and executing force-rich collaboration require compliance to human interaction. While compliance is often achieved with admittance control, many commercial robots lack the joint torque monitoring needed for such control. To address this challenge, we present an approach that uses tactile sensors and behavior cloning to transfer policies from robots with these capabilities to those without. We train a single policy that demonstrates positive transfer across embodiments, including robots without torque sensing. We demonstrate this positive transfer on four different tactile-enabled embodiments using the same policy trained on force-controlled robot data. Across multiple proposed metrics, the best performance came from a decomposed tactile shear-field representation combined with a pre-trained encoder, which improved success rates over alternative representations.

Sheng Zhong, Nima Fazeli, Dmitry Berenson

This paper presents Rummaging Using Mutual Information (RUMI), a method for online generation of robot action sequences to gather information about the pose of a known movable object in visually-occluded environments. Focusing on contact-rich rummaging, our approach leverages mutual information between the object pose distribution and robot trajectory for action planning. From an observed partial point cloud, RUMI deduces the compatible object pose distribution and approximates the mutual information of it with workspace occupancy in real time. Based on this, we develop an information gain cost function and a reachability cost function to keep the object within the robot's reach. These are integrated into a model predictive control (MPC) framework with a stochastic dynamics model, updating the pose distribution in a closed loop. Key contributions include a new belief framework for object pose estimation, an efficient information gain computation strategy, and a robust MPC-based control scheme. RUMI demonstrates superior performance in both simulated and real tasks compared to baseline methods.

Boyang Wang, Nikhil Sridhar, Chao Feng et al.

Clear, interpretable instructions are invaluable when attempting any complex task. Good instructions help to clarify the task and even anticipate the steps needed to solve it. In this work, we propose a robot learning framework for communicating, planning, and executing a wide range of tasks, dubbed This&That. This&That solves general tasks by leveraging video generative models, which, through training on internet-scale data, contain rich physical and semantic context. In this work, we tackle three fundamental challenges in video-based planning: 1) unambiguous task communication with simple human instructions, 2) controllable video generation that respects user intent, and 3) translating visual plans into robot actions. This&That uses language-gesture conditioning to generate video predictions, as a succinct and unambiguous alternative to existing language-only methods, especially in complex and uncertain environments. These video predictions are then fed into a behavior cloning architecture dubbed Diffusion Video to Action (DiVA), which outperforms prior state-of-the-art behavior cloning and video-based planning methods by substantial margins.

Yinpei Dai, Hongze Fu, Jayjun Lee et al.

Memory is critical for long-horizon and history-dependent robotic manipulation. Such tasks often involve counting repeated actions or manipulating objects that become temporarily occluded. Recent vision-language-action (VLA) models have begun to incorporate memory mechanisms; however, their evaluations remain confined to narrow, non-standardized settings. This limits their systematic understanding, comparison, and progress measurement. To address these challenges, we introduce RoboMME: a large-scale standardized benchmark for evaluating and advancing VLA models in long-horizon, history-dependent scenarios. Our benchmark comprises 16 manipulation tasks constructed under a carefully designed taxonomy that evaluates temporal, spatial, object, and procedural memory. We further develop a suite of 14 memory-augmented VLA variants built on the π0.5 backbone to systematically explore different memory representations across multiple integration strategies. Experimental results show that the effectiveness of memory representations is highly task-dependent, with each design offering distinct advantages and limitations across different tasks. Videos and code can be found at our website https://robomme.github.io.

Changhe Chen, Quantao Yang, Xiaohao Xu et al.

One of the central challenges preventing robots from acquiring complex manipulation skills is the prohibitive cost of collecting large-scale robot demonstrations. In contrast, humans are able to learn efficiently by watching others interact with their environment. To bridge this gap, we introduce semantic action flow as a core intermediate representation capturing the essential spatio-temporal manipulator-object interactions, invariant to superficial visual differences. We present ViSA-Flow, a framework that learns this representation self-supervised from unlabeled large-scale video data. First, a generative model is pre-trained on semantic action flows automatically extracted from large-scale human-object interaction video data, learning a robust prior over manipulation structure. Second, this prior is efficiently adapted to a target robot by fine-tuning on a small set of robot demonstrations processed through the same semantic abstraction pipeline. We demonstrate through extensive experiments on the CALVIN benchmark and real-world tasks that ViSA-Flow achieves state-of-the-art performance, particularly in low-data regimes, outperforming prior methods by effectively transferring knowledge from human video observation to robotic execution. Videos are available at https://visaflow-web.github.io/ViSAFLOW.

Basavasagar Patil, Sydney Belt, Jayjun Lee et al.

Increasingly large datasets of robot actions and sensory observations are being collected to train ever-larger neural networks. These datasets are collected based on tasks and while these tasks may be distinct in their descriptions, many involve very similar physical action sequences (e.g., 'pick up an apple' versus 'pick up an orange'). As a result, many datasets of robotic tasks are substantially imbalanced in terms of the physical robotic actions they represent. In this work, we propose a simple sampling strategy for policy training that mitigates this imbalance. Our method requires only a few lines of code to integrate into existing codebases and improves generalization. We evaluate our method in both pre-training small models and fine-tuning large foundational models. Our results show substantial improvements on low-resource tasks compared to prior state-of-the-art methods, without degrading performance on high-resource tasks. This enables more effective use of model capacity for multi-task policies. We also further validate our approach in a real-world setup on a Franka Panda robot arm across a diverse set of tasks.

Samanta Rodriguez, Yiming Dou, Miquel Oller et al.

Today's visuo-tactile sensors come in many shapes and sizes, making it challenging to develop general-purpose tactile representations. This is because most models are tied to a specific sensor design. To address this challenge, we propose two approaches to cross-sensor image generation. The first is an end-to-end method that leverages paired data (Touch2Touch). The second method builds an intermediate depth representation and does not require paired data (T2D2: Touch-to-Depth-to-Touch). Both methods enable the use of sensor-specific models across multiple sensors via the cross-sensor touch generation process. Together, these models offer flexible solutions for sensor translation, depending on data availability and application needs. We demonstrate their effectiveness on downstream tasks such as in-hand pose estimation and behavior cloning, successfully transferring models trained on one sensor to another. Project page: https://samantabelen.github.io/cross_sensor_touch_generation.

Jayjun Lee, Nima Fazeli

Mastering dexterous, contact-rich object manipulation demands precise estimation of both in-hand object poses and external contact locations$\unicode{x2013}$tasks particularly challenging due to partial and noisy observations. We present ViTaSCOPE: Visuo-Tactile Simultaneous Contact and Object Pose Estimation, an object-centric neural implicit representation that fuses vision and high-resolution tactile feedback. By representing objects as signed distance fields and distributed tactile feedback as neural shear fields, ViTaSCOPE accurately localizes objects and registers extrinsic contacts onto their 3D geometry as contact fields. Our method enables seamless reasoning over complementary visuo-tactile cues by leveraging simulation for scalable training and zero-shot transfers to the real-world by bridging the sim-to-real gap. We evaluate our method through comprehensive simulated and real-world experiments, demonstrating its capabilities in dexterous manipulation scenarios.

Mark Van der Merwe, Youngsun Wi, Dmitry Berenson et al.

Reasoning over the interplay between object deformation and force transmission through contact is central to the manipulation of compliant objects. In this paper, we propose Neural Deforming Contact Field (NDCF), a representation that jointly models object deformations and contact patches from visuo-tactile feedback using implicit representations. Representing the object geometry and contact with the environment implicitly allows a single model to predict contact patches of varying complexity. Additionally, learning geometry and contact simultaneously allows us to enforce physical priors, such as ensuring contacts lie on the surface of the object. We propose a neural network architecture to learn a NDCF, and train it using simulated data. We then demonstrate that the learned NDCF transfers directly to the real-world without the need for fine-tuning. We benchmark our proposed approach against a baseline representing geometry and contact patches with point clouds. We find that NDCF performs better on simulated data and in transfer to the real-world.

Sheng Zhong, Nima Fazeli, Dmitry Berenson

This paper proposes a novel method for estimating the set of plausible poses of a rigid object from a set of points with volumetric information, such as whether each point is in free space or on the surface of the object. In particular, we study how pose can be estimated from force and tactile data arising from contact. Using data derived from contact is challenging because it is inherently less information-dense than visual data, and thus the pose estimation problem is severely under-constrained when there are few contacts. Rather than attempting to estimate the true pose of the object, which is not tractable without a large number of contacts, we seek to estimate a plausible set of poses which obey the constraints imposed by the sensor data. Existing methods struggle to estimate this set because they are either designed for single pose estimates or require informative priors to be effective. Our approach to this problem, Constrained pose Hypothesis Set Elimination (CHSEL), has three key attributes: 1) It considers volumetric information, which allows us to account for known free space; 2) It uses a novel differentiable volumetric cost function to take advantage of powerful gradient-based optimization tools; and 3) It uses methods from the Quality Diversity (QD) optimization literature to produce a diverse set of high-quality poses. To our knowledge, QD methods have not been used previously for pose registration. We also show how to update our plausible pose estimates online as more data is gathered by the robot. Our experiments suggest that CHSEL shows large performance improvements over several baseline methods for both simulated and real-world data.

Youngsun Wi, Pete Florence, Andy Zeng et al.

Deformable object manipulation requires computationally efficient representations that are compatible with robotic sensing modalities. In this paper, we present VIRDO:an implicit, multi-modal, and continuous representation for deformable-elastic objects. VIRDO operates directly on visual (point cloud) and tactile (reaction forces) modalities and learns rich latent embeddings of contact locations and forces to predict object deformations subject to external contacts.Here, we demonstrate VIRDOs ability to: i) produce high-fidelity cross-modal reconstructions with dense unsupervised correspondences, ii) generalize to unseen contact formations,and iii) state-estimation with partial visio-tactile feedback

Sheng Zhong, Nima Fazeli, Dmitry Berenson

Retrieving an object from cluttered spaces suchas cupboards, refrigerators, or bins requires tracking objects with limited or no visual sensing. In these scenarios, contact feedback is necessary to estimate the pose of the objects, yet the objects are movable while their shapes and number may be unknown, making the association of contacts with objects extremely difficult. While previous work has focused on multi-target tracking, the assumptions therein prohibit using prior methods given only the contact-sensing modality. Instead, this paper proposes the method Soft Tracking Using Contacts for Cluttered Objects (STUCCO) that tracks the belief over contact point locations and implicit object associations using a particle filter. This method allows ambiguous object associations of past contacts to be revised as new information becomes available. We apply STUCCO to the Blind Object Retrieval problem, where a target object of known shape but unknown pose must be retrieved from clutter. Our results suggest that our method outperforms baselines in four simulation environments, and on a real robot, where contact sensing is noisy. In simulation, we achieve grasp success of at least 65% on all environments while no baselines achieve over 5%.

Sheng Zhong, Zhenyuan Zhang, Nima Fazeli et al.

We propose an approach to online model adaptation and control in the challenging case of hybrid and discontinuous dynamics where actions may lead to difficult-to-escape "trap" states, under a given controller. We first learn dynamics for a system without traps from a randomly collected training set (since we do not know what traps will be encountered online). These "nominal" dynamics allow us to perform tasks in scenarios where the dynamics matches the training data, but when unexpected traps arise in execution, we must find a way to adapt our dynamics and control strategy and continue attempting the task. Our approach, Trap-Aware Model Predictive Control (TAMPC), is a two-level hierarchical control algorithm that reasons about traps and non-nominal dynamics to decide between goal-seeking and recovery policies. An important requirement of our method is the ability to recognize nominal dynamics even when we encounter data that is out-of-distribution w.r.t the training data. We achieve this by learning a representation for dynamics that exploits invariance in the nominal environment, thus allowing better generalization. We evaluate our method on simulated planar pushing and peg-in-hole as well as real robot peg-in-hole problems against adaptive control, reinforcement learning, trap-handling baselines, where traps arise due to unexpected obstacles that we only observe through contact. Our results show that our method outperforms the baselines on difficult tasks, and is comparable to prior trap-handling methods on easier tasks.

Nima Fazeli, Anurag Ajay, Alberto Rodriguez

The ability to simulate and predict the outcome of contacts is paramount to the successful execution of many robotic tasks. Simulators are powerful tools for the design of robots and their behaviors, yet the discrepancy between their predictions and observed data limit their usability. In this paper, we propose a self-supervised approach to learning residual models for rigid-body simulators that exploits corrections of contact models to refine predictive performance and propagate uncertainty. We empirically evaluate the framework by predicting the outcomes of planar dice rolls and compare it's performance to state-of-the-art techniques.

Anurag Ajay, Maria Bauza, Jiajun Wu et al.

Physics engines play an important role in robot planning and control; however, many real-world control problems involve complex contact dynamics that cannot be characterized analytically. Most physics engines therefore employ . approximations that lead to a loss in precision. In this paper, we propose a hybrid dynamics model, simulator-augmented interaction networks (SAIN), combining a physics engine with an object-based neural network for dynamics modeling. Compared with existing models that are purely analytical or purely data-driven, our hybrid model captures the dynamics of interacting objects in a more accurate and data-efficient manner.Experiments both in simulation and on a real robot suggest that it also leads to better performance when used in complex control tasks. Finally, we show that our model generalizes to novel environments with varying object shapes and materials.

Anurag Ajay, Jiajun Wu, Nima Fazeli et al.

An efficient, generalizable physical simulator with universal uncertainty estimates has wide applications in robot state estimation, planning, and control. In this paper, we build such a simulator for two scenarios, planar pushing and ball bouncing, by augmenting an analytical rigid-body simulator with a neural network that learns to model uncertainty as residuals. Combining symbolic, deterministic simulators with learnable, stochastic neural nets provides us with expressiveness, efficiency, and generalizability simultaneously. Our model outperforms both purely analytical and purely learned simulators consistently on real, standard benchmarks. Compared with methods that model uncertainty using Gaussian processes, our model runs much faster, generalizes better to new object shapes, and is able to characterize the complex distribution of object trajectories.

Nima Fazeli, Samuel Zapolsky, Evan Drumwright et al.

In this paper we demonstrate the limitations of common rigid-body contact models used in the robotics community by comparing them to a collection of data-driven and data-reinforced models that exploit underlying structure inspired by the rigid contact paradigm. We evaluate and compare the analytical and data-driven contact models on an empirical planar impact data-set, and show that the learned models are able to outperform their analytical counterparts with a small training set.

Nima Fazeli, Samuel Zapolsky, Evan Drumwright et al.

The ability to reason about and predict the outcome of contacts is paramount to the successful execution of many robot tasks. Analytical rigid-body contact models are used extensively in planning and control due to their computational efficiency and simplicity, yet despite their prevalence, little if any empirical comparison of these models has been made and it is unclear how well they approximate contact outcomes. In this paper, we first formulate a system identification approach for six commonly used contact models in the literature, and use the proposed method to find parameters for an experimental data-set of impacts. Next, we compare the models empirically, and establish a task specific upper bound on the performance of the models and the rigid-body contact model paradigm. We highlight the limitations of these models, salient failure modes, and the care that should be taken in parameter selection, which are ultimately difficult to give a physical interpretation.

Andy Zeng, Shuran Song, Kuan-Ting Yu et al.

This paper presents a robotic pick-and-place system that is capable of grasping and recognizing both known and novel objects in cluttered environments. The key new feature of the system is that it handles a wide range of object categories without needing any task-specific training data for novel objects. To achieve this, it first uses a category-agnostic affordance prediction algorithm to select and execute among four different grasping primitive behaviors. It then recognizes picked objects with a cross-domain image classification framework that matches observed images to product images. Since product images are readily available for a wide range of objects (e.g., from the web), the system works out-of-the-box for novel objects without requiring any additional training data. Exhaustive experimental results demonstrate that our multi-affordance grasping achieves high success rates for a wide variety of objects in clutter, and our recognition algorithm achieves high accuracy for both known and novel grasped objects. The approach was part of the MIT-Princeton Team system that took 1st place in the stowing task at the 2017 Amazon Robotics Challenge. All code, datasets, and pre-trained models are available online at http://arc.cs.princeton.edu

Kuan-Ting Yu, Maria Bauza, Nima Fazeli et al.

Pushing is a motion primitive useful to handle objects that are too large, too heavy, or too cluttered to be grasped. It is at the core of much of robotic manipulation, in particular when physical interaction is involved. It seems reasonable then to wish for robots to understand how pushed objects move. In reality, however, robots often rely on approximations which yield models that are computable, but also restricted and inaccurate. Just how close are those models? How reasonable are the assumptions they are based on? To help answer these questions, and to get a better experimental understanding of pushing, we present a comprehensive and high-fidelity dataset of planar pushing experiments. The dataset contains timestamped poses of a circular pusher and a pushed object, as well as forces at the interaction.We vary the push interaction in 6 dimensions: surface material, shape of the pushed object, contact position, pushing direction, pushing speed, and pushing acceleration. An industrial robot automates the data capturing along precisely controlled position-velocity-acceleration trajectories of the pusher, which give dense samples of positions and forces of uniform quality. We finish the paper by characterizing the variability of friction, and evaluating the most common assumptions and simplifications made by models of frictional pushing in robotics.

Kuan-Ting Yu, Nima Fazeli, Nikhil Chavan-Dafle et al.

The Amazon Picking Challenge (APC), held alongside the International Conference on Robotics and Automation in May 2015 in Seattle, challenged roboticists from academia and industry to demonstrate fully automated solutions to the problem of picking objects from shelves in a warehouse fulfillment scenario. Packing density, object variability, speed, and reliability are the main complexities of the task. The picking challenge serves both as a motivation and an instrument to focus research efforts on a specific manipulation problem. In this document, we describe Team MIT's approach to the competition, including design considerations, contributions, and performance, and we compile the lessons learned. We also describe what we think are the main remaining challenges.