SEARCH · Search arXiv

Results for “robot learning”

Search indexed arXiv papers on artificial intelligence, large language models, computer vision and robotics. Read source abstracts and follow links to arXiv.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 1,477 records · Page 82Linked to original sources

arXiv2018

Grasp2Vec: Learning Object Representations from Self-Supervised Grasping

Well structured visual representations can make robot learning faster and can improve generalization. In this paper, we study how we can acquire effective object-centric representations for robotic manipulation tasks without human labeling by using autonomous robot interaction with the environment. Such representation learning methods can benefit from continuous refinement of the representation as the robot collects more experience, allowing them to scale effectively without human intervention. Our representation learning approach is based on object persistence: when a robot removes an object from a scene, the representation of that scene should change according to the features of the object that was removed. We formulate an arithmetic relationship between feature vectors from this observation, and use it to learn a representation of scenes and objects that can then be used to identify object instances, localize them in the scene, and perform goal-directed grasping tasks where the robot must retrieve commanded objects from a bin. The same grasping procedure can also be used to automatically collect training data for our method, by recording images of scenes, grasping and removing an object, and recording the outcome. Our experiments demonstrate that this self-supervised approach for tasked grasping substantially outperforms direct reinforcement learning from images and prior representation learning methods.

cs.RO↗

arXiv2022

Continual Learning for Affective Robotics: A Proof of Concept for Wellbeing

Sustaining real-world human-robot interactions requires robots to be sensitive to human behavioural idiosyncrasies and adapt their perception and behaviour models to cater to these individual preferences. For affective robots, this entails learning to adapt to individual affective behaviour to offer a personalised interaction experience to each individual. Continual Learning (CL) has been shown to enable real-time adaptation in agents, allowing them to learn with incrementally acquired data while preserving past knowledge. In this work, we present a novel framework for real-world application of CL for modelling personalised human-robot interactions using a CL-based affect perception mechanism. To evaluate the proposed framework, we undertake a proof-of-concept user study with 20 participants interacting with the Pepper robot using three variants of interaction behaviour: static and scripted, using affect-based adaptation without personalisation, and using affect-based adaptation with continual personalisation. Our results demonstrate a clear preference in the participants for CL-based continual personalisation with significant improvements observed in the robot's anthropomorphism, animacy and likeability ratings as well as the interactions being rated significantly higher for warmth and comfort as the robot is rated as significantly better at understanding how the participants feel.

cs.HC↗

arXiv2019

Long-Term Personalization of an In-Home Socially Assistive Robot for Children With Autism Spectrum Disorders

Socially assistive robots (SAR) have shown great potential to augment the social and educational development of children with autism spectrum disorders (ASD). As SAR continues to substantiate itself as an effective enhancement to human intervention, researchers have sought to study its longitudinal impacts in real-world environments, including the home. Computational personalization stands out as a central computational challenge as it is necessary to enable SAR systems to adapt to each child's unique and changing needs. Toward that end, we formalized personalization as a hierarchical human robot learning framework (hHRL) consisting of five controllers (disclosure, promise, instruction, feedback, and inquiry) mediated by a meta-controller that utilized reinforcement learning to personalize instruction challenge levels and robot feedback based on each user's unique learning patterns. We instantiated and evaluated the approach in a study with 17 children with ASD, aged 3 to 7 years old, over month-long interventions in their homes. Our findings demonstrate that the fully autonomous SAR system was able to personalize its instruction and feedback over time to each child's proficiency. As a result, every child participant showed improvements in targeted skills and long-term retention of intervention content. Moreover, all child users were engaged for a majority of the intervention, and their families reported the SAR system to be useful and adaptable. In summary, our results show that autonomous, personalized SAR interventions are both feasible and effective in providing long-term in-home developmental support for children with diverse learning needs.

cs.RO↗

arXiv2022

Adaptive actuation of magnetic soft robots using deep reinforcement learning

Magnetic soft robots have attracted growing interest due to their unique advantages in terms of untethered actuation and excellent controllability. However, finding the required magnetization patterns or magnetic fields to achieve the desired functions of these robots is quite challenging in many cases. No unified framework for design has been proposed yet, and existing methods mainly rely on manual heuristics, which are hard to satisfy the high complexity level of the desired robotic motion. Here, we develop an intelligent method to solve the related inverse-design problems, implemented by introducing a novel simulation platform for magnetic soft robots based on Cosserat rod models and a deep reinforcement learning framework based on TD3. We demonstrate that magnetic soft robots with different magnetization patterns can learn to move without human guidance in simulations, and effective magnetic fields can be autonomously generated that can then be applied directly to real magnetic soft robots in an open-loop way.

cs.RO↗

arXiv2026

COFFAIL: A Dataset of Successful and Anomalous Robot Skill Executions in the Context of Coffee Preparation

In the context of robot learning for manipulation, curated datasets are an important resource for advancing the state of the art; however, available datasets typically only include successful executions or are focused on one particular type of skill. In this short paper, we briefly describe a dataset of various skills performed in the context of coffee preparation. The dataset, which we call COFFAIL, includes both successful and anomalous skill execution episodes collected with a physical robot in a kitchen environment, a couple of which are performed with bimanual manipulation. In addition to describing the data collection setup and the collected data, the paper illustrates the use of the data in COFFAIL to learn a robot policy using imitation learning.

cs.RO↗

arXiv2025

DynaRend: Learning 3D Dynamics via Masked Future Rendering for Robotic Manipulation

Learning generalizable robotic manipulation policies remains a key challenge due to the scarcity of diverse real-world training data. While recent approaches have attempted to mitigate this through self-supervised representation learning, most either rely on 2D vision pretraining paradigms such as masked image modeling, which primarily focus on static semantics or scene geometry, or utilize large-scale video prediction models that emphasize 2D dynamics, thus failing to jointly learn the geometry, semantics, and dynamics required for effective manipulation. In this paper, we present DynaRend, a representation learning framework that learns 3D-aware and dynamics-informed triplane features via masked reconstruction and future prediction using differentiable volumetric rendering. By pretraining on multi-view RGB-D video data, DynaRend jointly captures spatial geometry, future dynamics, and task semantics in a unified triplane representation. The learned representations can be effectively transferred to downstream robotic manipulation tasks via action value map prediction. We evaluate DynaRend on two challenging benchmarks, RLBench and Colosseum, as well as in real-world robotic experiments, demonstrating substantial improvements in policy success rate, generalization to environmental perturbations, and real-world applicability across diverse manipulation tasks.

cs.RO↗

arXiv2025

Consensus Building in Human-robot Co-learning via Bias Controlled Nonlinear Opinion Dynamics and Non-verbal Communication through Robotic Eyes

Consensus between humans and robots is crucial as robotic agents become more prevalent and deeply integrated into our daily lives. This integration presents both unprecedented opportunities and notable challenges for effective collaboration. However, the active guidance of human actions and their integration in co-learning processes, where humans and robots mutually learn from each other, remains under-explored. This article demonstrates how consensus between human and robot opinions can be established by modeling decision-making processes as non-linear opinion dynamics. We utilize dynamic bias as a control parameter to steer the robot's opinion toward consensus and employ visual cues via a robotic eye gaze to guide human decisions. These non-verbal cues communicate the robot's future intentions, gradually guiding human decisions to align with them. To design robot behavior for consensus, we integrate a human opinion observation algorithm with the robot's opinion formation, controlling its actions based on that formed opinion. Experiments with $51$ participants ($N=51$) in a two-choice decision-making task show that effective consensus and trust can be established in a human--robot co-learning setting by guiding human decisions through nonverbal robotic cues and using bias-controlled opinion dynamics to shape robot behavior. Finally, we provide detailed information on the perceived cognitive load and the behavior of robotic eyes based on user feedback and post-experiment interviews.

cs.RO↗

arXiv2026

Supervised Mixture-of-Experts for Surgical Grasping and Retraction

Imitation learning has achieved remarkable success in robotic manipulation, yet its application to surgical robotics remains challenging due to data scarcity, constrained workspaces, and the need for an exceptional level of safety and predictability. We present a supervised Mixture-of-Experts (MoE) architecture designed for phase-structured surgical manipulation tasks, which can be added on top of any autonomous policy. Unlike prior surgical robot learning approaches that rely on multi-camera setups or thousands of demonstrations, we show that a lightweight action decoder policy like Action Chunking Transformer (ACT) can learn complex, long-horizon manipulation from less than 150 demonstrations using solely stereo endoscopic images, when equipped with our architecture. We evaluate our approach on the collaborative surgical task of bowel grasping and retraction, where a robot assistant interprets visual cues from a human surgeon, executes targeted grasping on deformable tissue, and performs sustained retraction. Our results show that generalist Vision Language Action models fail to acquire the task entirely, even under standard in-distribution conditions. Furthermore, while standard ACT achieves moderate success in-distribution, adopting a supervised MoE architecture significantly boosts its performance, yielding higher success rates in-distribution and demonstrating superior robustness in out-of-distribution scenarios, including novel grasp locations, reduced illumination, and partial occlusions. Notably, it generalizes to unseen testing viewpoints and also transfers zero-shot to ex vivo porcine tissue without additional training, offering a promising pathway toward in vivo deployment. To support this statement, we present qualitative preliminary results of policy roll-outs during in vivo porcine surgery.

cs.RO↗

arXiv2019

Continuous Value Iteration (CVI) Reinforcement Learning and Imaginary Experience Replay (IER) for learning multi-goal, continuous action and state space controllers

This paper presents a novel model-free Reinforcement Learning algorithm for learning behavior in continuous action, state, and goal spaces. The algorithm approximates optimal value functions using non-parametric estimators. It is able to efficiently learn to reach multiple arbitrary goals in deterministic and nondeterministic environments. To improve generalization in the goal space, we propose a novel sample augmentation technique. Using these methods, robots learn faster and overall better controllers. We benchmark the proposed algorithms using simulation and a real-world voltage controlled robot that learns to maneuver in a non-observable Cartesian task space.

cs.AI↗

arXiv2022

Design strategies for controlling neuron-connected robots using reinforcement learning

Despite the growing interest in robot control utilizing the computation of biological neurons, context-dependent behavior by neuron-connected robots remains a challenge. Context-dependent behavior here is defined as behavior that is not the result of a simple sensory-motor coupling, but rather based on an understanding of the task goal. This paper proposes design principles for training neuron-connected robots based on task goals to achieve context-dependent behavior. First, we employ deep reinforcement learning (RL) to enable training that accounts for goal achievements. Second, we propose a neuron simulator as a probability distribution based on recorded neural data, aiming to represent physiologically valid neural dynamics while avoiding complex modeling with high computational costs. Furthermore, we propose to update the simulators during the training to bridge the gap between the simulation and the real settings. The experiments showed that the robot gradually learned context-dependent behaviors in pole balancing and robot navigation tasks. Moreover, the learned policies were valid for neural simulators based on novel neural data, and the task performance increased by updating the simulators during training. These results suggest the effectiveness of the proposed design principle for the context-dependent behavior of neuron-connected robots.

cs.RO↗

arXiv2025

Robotic Paper Wrapping by Learning Force Control

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.

cs.RO↗

arXiv2024

CelluloTactix: Towards Empowering Collaborative Online Learning through Tangible Haptic Interaction with Cellulo Robots

Online learning has soared in popularity in the educational landscape of COVID-19 and carries the benefits of increased flexibility and access to far-away training resources. However, it also restricts communication between peers and teachers, limits physical interactions and confines learning to the computer screen and keyboard. In this project, we designed a novel way to engage students in collaborative online learning by using haptic-enabled tangible robots, Cellulo. We built a library which connects two robots remotely for a learning activity based around the structure of a biological cell. To discover how separate modes of haptic feedback might differentially affect collaboration, two modes of haptic force-feedback were implemented (haptic co-location and haptic consensus). With a case study, we found that the haptic co-location mode seemed to stimulate collectivist behaviour to a greater extent than the haptic consensus mode, which was associated with individualism and less interaction. While the haptic co-location mode seemed to encourage information pooling, participants using the haptic consensus mode tended to focus more on technical co-ordination. This work introduces a novel system that can provide interesting insights on how to integrate haptic feedback into collaborative remote learning activities in future.

cs.HC↗

arXiv2018

A Survey of Deep Network Solutions for Learning Control in Robotics: From Reinforcement to Imitation

Deep learning techniques have been widely applied, achieving state-of-the-art results in various fields of study. This survey focuses on deep learning solutions that target learning control policies for robotics applications. We carry out our discussions on the two main paradigms for learning control with deep networks: deep reinforcement learning and imitation learning. For deep reinforcement learning (DRL), we begin from traditional reinforcement learning algorithms, showing how they are extended to the deep context and effective mechanisms that could be added on top of the DRL algorithms. We then introduce representative works that utilize DRL to solve navigation and manipulation tasks in robotics. We continue our discussion on methods addressing the challenge of the reality gap for transferring DRL policies trained in simulation to real-world scenarios, and summarize robotics simulation platforms for conducting DRL research. For imitation leaning, we go through its three main categories, behavior cloning, inverse reinforcement learning and generative adversarial imitation learning, by introducing their formulations and their corresponding robotics applications. Finally, we discuss the open challenges and research frontiers.

cs.RO↗

arXiv2022

On Decentralizing Federated Reinforcement Learning in Multi-Robot Scenarios

Federated Learning (FL) allows for collaboratively aggregating learned information across several computing devices and sharing the same amongst them, thereby tackling issues of privacy and the need of huge bandwidth. FL techniques generally use a central server or cloud for aggregating the models received from the devices. Such centralized FL techniques suffer from inherent problems such as failure of the central node and bottlenecks in channel bandwidth. When FL is used in conjunction with connected robots serving as devices, a failure of the central controlling entity can lead to a chaotic situation. This paper describes a mobile agent based paradigm to decentralize FL in multi-robot scenarios. Using Webots, a popular free open-source robot simulator, and Tartarus, a mobile agent platform, we present a methodology to decentralize federated learning in a set of connected robots. With Webots running on different connected computing systems, we show how mobile agents can perform the task of Decentralized Federated Reinforcement Learning (dFRL). Results obtained from experiments carried out using Q-learning and SARSA by aggregating their corresponding Q-tables, show the viability of using decentralized FL in the domain of robotics. Since the proposed work can be used in conjunction with other learning algorithms and also real robots, it can act as a vital tool for the study of decentralized FL using heterogeneous learning algorithms concurrently in multi-robot scenarios.

cs.RO↗

arXiv2025

TD-GRPC: Temporal Difference Learning with Group Relative Policy Constraint for Humanoid Locomotion

Robot learning in high-dimensional control settings, such as humanoid locomotion, presents persistent challenges for reinforcement learning (RL) algorithms due to unstable dynamics, complex contact interactions, and sensitivity to distributional shifts during training. Model-based methods, \textit{e.g.}, Temporal-Difference Model Predictive Control (TD-MPC), have demonstrated promising results by combining short-horizon planning with value-based learning, enabling efficient solutions for basic locomotion tasks. However, these approaches remain ineffective in addressing policy mismatch and instability introduced by off-policy updates. Thus, in this work, we introduce Temporal-Difference Group Relative Policy Constraint (TD-GRPC), an extension of the TD-MPC framework that unifies Group Relative Policy Optimization (GRPO) with explicit Policy Constraints (PC). TD-GRPC applies a trust-region constraint in the latent policy space to maintain consistency between the planning priors and learned rollouts, while leveraging group-relative ranking to assess and preserve the physical feasibility of candidate trajectories. Unlike prior methods, TD-GRPC achieves robust motions without modifying the underlying planner, enabling flexible planning and policy learning. We validate our method across a locomotion task suite ranging from basic walking to highly dynamic movements on the 26-DoF Unitree H1-2 humanoid robot. Through simulation results, TD-GRPC demonstrates its improvements in stability and policy robustness with sampling efficiency while training for complex humanoid control tasks.

cs.RO↗

arXiv2021

Unsupervised Lexical Acquisition of Relative Spatial Concepts Using Spoken User Utterances

This paper proposes methods for unsupervised lexical acquisition for relative spatial concepts using spoken user utterances. A robot with a flexible spoken dialog system must be able to acquire linguistic representation and its meaning specific to an environment through interactions with humans as children do. Specifically, relative spatial concepts (e.g., front and right) are widely used in our daily lives, however, it is not obvious which object is a reference object when a robot learns relative spatial concepts. Therefore, we propose methods by which a robot without prior knowledge of words can learn relative spatial concepts. The methods are formulated using a probabilistic model to estimate the proper reference objects and distributions representing concepts simultaneously. The experimental results show that relative spatial concepts and a phoneme sequence representing each concept can be learned under the condition that the robot does not know which located object is the reference object. Additionally, we show that two processes in the proposed method improve the estimation accuracy of the concepts: generating candidate word sequences by class n-gram and selecting word sequences using location information. Furthermore, we show that clues to reference objects improve accuracy even though the number of candidate reference objects increases.

cs.AI↗

arXiv2021

Manipulator-Independent Representations for Visual Imitation

Imitation learning is an effective tool for robotic learning tasks where specifying a reinforcement learning (RL) reward is not feasible or where the exploration problem is particularly difficult. Imitation, typically behavior cloning or inverse RL, derive a policy from a collection of first-person action-state trajectories. This is contrary to how humans and other animals imitate: we observe a behavior, even from other species, understand its perceived effect on the state of the environment, and figure out what actions our body can perform to reach a similar outcome. In this work, we explore the possibility of third-person visual imitation of manipulation trajectories, only from vision and without access to actions, demonstrated by embodiments different to the ones of our imitating agent. Specifically, we investigate what would be an appropriate representation method with which an RL agent can visually track trajectories of complex manipulation behavior -- non-planar with multiple-object interactions -- demonstrated by experts with different embodiments. We present a way to train manipulator-independent representations (MIR) that primarily focus on the change in the environment and have all the characteristics that make them suitable for cross-embodiment visual imitation with RL: cross-domain alignment, temporal smoothness, and being actionable. We show that with our proposed method our agents are able to imitate, with complex robot control, trajectories from a variety of embodiments and with significant visual and dynamics differences, e.g. simulation-to-reality gap.

cs.RO↗

arXiv2024

Survey on safe robot control via learning

Control systems are critical to modern technological infrastructure, spanning industries from aerospace to healthcare. This survey explores the landscape of safe robot learning, investigating methods that balance high-performance control with rigorous safety constraints. By examining classical control techniques, learning-based approaches, and embedded system design, the research seeks to understand how robotic systems can be developed to prevent hazardous states while maintaining optimal performance across complex operational environments.

cs.RO↗