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 757 records · Page 42Linked to original sources

arXiv2026

NFPO: Stabilized Policy Optimization of Normalizing Flow for Robotic Policy Learning

Deep Reinforcement Learning (DRL) has experienced significant advancements in recent years and has been widely used in many fields. In DRL-based robotic policy learning, however, current de facto policy parameterization is still multivariate Gaussian (with diagonal covariance matrix), which lacks the ability to model multi-modal distribution. In this work, we explore the adoption of a modern network architecture, i.e. Normalizing Flow (NF) as the policy parameterization for its ability of multi-modal modeling, closed form of log probability and low computation and memory overhead. However, naively training NF in online Reinforcement Learning (RL) usually leads to training instability. We provide a detailed analysis for this phenomenon and successfully address it via simple but effective technique. With extensive experiments in multiple simulation environments, we show our method, NFPO could obtain robust and strong performance in widely used robotic learning tasks and successfully transfer into real-world robots.

cs.RO↗

arXiv2022

Deep Causal Learning for Robotic Intelligence

This invited review discusses causal learning in the context of robotic intelligence. The paper introduced the psychological findings on causal learning in human cognition, then it introduced the traditional statistical solutions on causal discovery and causal inference. The paper reviewed recent deep causal learning algorithms with a focus on their architectures and the benefits of using deep nets and discussed the gap between deep causal learning and the needs of robotic intelligence.

cs.RO↗

arXiv2025

No Need for Real 3D: Fusing 2D Vision with Pseudo 3D Representations for Robotic Manipulation Learning

Recently,vision-based robotic manipulation has garnered significant attention and witnessed substantial advancements. 2D image-based and 3D point cloud-based policy learning represent two predominant paradigms in the field, with recent studies showing that the latter consistently outperforms the former in terms of both policy performance and generalization, thereby underscoring the value and significance of 3D information. However, 3D point cloud-based approaches face the significant challenge of high data acquisition costs, limiting their scalability and real-world deployment. To address this issue, we propose a novel framework NoReal3D: which introduces the 3DStructureFormer, a learnable 3D perception module capable of transforming monocular images into geometrically meaningful pseudo-point cloud features, effectively fused with the 2D encoder output features. Specially, the generated pseudo-point clouds retain geometric and topological structures so we design a pseudo-point cloud encoder to preserve these properties, making it well-suited for our framework. We also investigate the effectiveness of different feature fusion strategies.Our framework enhances the robot's understanding of 3D spatial structures while completely eliminating the substantial costs associated with 3D point cloud acquisition.Extensive experiments across various tasks validate that our framework can achieve performance comparable to 3D point cloud-based methods, without the actual point cloud data.

cs.RO↗

arXiv2026

Where-to-Learn: Analytical Policy Gradient Directed Exploration for On-Policy Robotic Reinforcement Learning

On-policy reinforcement learning (RL) algorithms have demonstrated great potential in robotic control, where effective exploration is crucial for efficient and high-quality policy learning. However, how to encourage the agent to explore the better trajectories efficiently remains a challenge. Most existing methods incentivize exploration by maximizing the policy entropy or encouraging novel state visiting regardless of the potential state value. We propose a new form of directed exploration that uses analytical policy gradients from a differentiable dynamics model to inject task-aware, physics-guided guidance, thereby steering the agent towards high-reward regions for accelerated and more effective policy learning.

cs.RO↗

arXiv2022

Safety-Critical Learning of Robot Control with Temporal Logic Specifications

Reinforcement learning (RL) is a promising approach. However, success is limited to real-world applications, because ensuring safe exploration and facilitating adequate exploitation is a challenge for controlling robotic systems with unknown models and measurement uncertainties. The learning problem becomes even more difficult for complex tasks over continuous state-action. In this paper, we propose a learning-based robotic control framework consisting of several aspects: (1) we leverage Linear Temporal Logic (LTL) to express complex tasks over infinite horizons that are translated to a novel automaton structure; (2) we detail an innovative reward scheme for LTL satisfaction with a probabilistic guarantee. Then, by applying a reward shaping technique, we develop a modular policy-gradient architecture exploiting the benefits of the automaton structure to decompose overall tasks and enhance the performance of learned controllers; (3) by incorporating Gaussian Processes (GPs) to estimate the uncertain dynamic systems, we synthesize a model-based safe exploration during the learning process using Exponential Control Barrier Functions (ECBFs) that generalize systems with high-order relative degrees; (4) to further improve the efficiency of exploration, we utilize the properties of LTL automata and ECBFs to propose a safe guiding process. Finally, we demonstrate the effectiveness of the framework via several robotic environments. We show an ECBF-based modular deep RL algorithm that achieves near-perfect success rates and safety guarding with high probability confidence during training.

cs.RO↗

arXiv2020

Adaptive Neural Trajectory Tracking Control for Flexible-Joint Robots with Online Learning

Collaborative robots and space manipulators contain significant joint flexibility. It complicates the control design, compromises the control bandwidth, and limits the tracking accuracy. The imprecise knowledge of the flexible joint dynamics compounds the challenge. In this paper, we present a new control architecture for controlling flexible-joint robots. Our approach uses a multi-layer neural network to approximate unknown dynamics needed for the feedforward control. The network may be viewed as a linear-in-parameter representation of the robot dynamics, with the nonlinear basis of the robot dynamics connected to the linear output layer. The output layer weights are updated based on the tracking error and the nonlinear basis. The internal weights of the nonlinear basis are updated by online backpropagation to further reduce the tracking error. To use time scale separation to reduce the coupling of the two steps - the update of the internal weights is at a lower rate compared to the update of the output layer weights. With the update of the output layer weights, our controller adapts quickly to the unknown dynamics change and disturbances (such as attaching a load). The update of the internal weights would continue to improve the converge of the nonlinear basis functions. We show the stability of the proposed scheme under the "outer loop" control, where the commanded joint position is considered as the control input. Simulation and physical experiments are conducted to demonstrate the performance of the proposed controller on a Baxter robot, which exhibits significant joint flexibility due to the series-elastic joint actuators.

cs.RO↗

arXiv2026

Temporally Centered SIGReg Improves LeWorldModel Representations for Robot Policy Learning

Recent work on LeWorldModel (LeWM) has shown that the Sketched Isotropic Gaussian Regularizer (SIGReg) enables stable end-to-end world model learning from pixels by regularizing the latent representation toward an isotropic Gaussian. While effective for latent-space planning, the representations learned by Raw LeWM are poorly suited for downstream robot policy learning. In this paper, through Monte Carlo analysis, we show that the Raw LeWM objective biases variance allocation toward the temporally persistent component, thereby suppressing the variance of the temporally centered residual. Consistent with this analysis, trained Raw LeWM representations exhibit suppressed residual variation and reduced decodability of robot state and dynamics, particularly gripper dynamics, which are crucial for robotic manipulation. To address this issue, we apply SIGReg to temporally centered residuals rather than to the whole latent representation. This simple change decouples persistent and residual variance allocation while retaining an effective anti-collapse property. On the LIBERO benchmark, our method improves downstream policy success on the Goal suite by 1.66x and raises the average success rate across all suites from 63.6% to 83.8%. Without external pretraining, it also outperforms both Diffusion Policy trained from scratch and the pretrained OpenVLA baseline. These results associate the variance-allocation bias of Raw LeWM with the downstream policy gap, and show that decoupling persistent and residual variation yields representations better suited for downstream robot policy learning.

cs.LG↗

arXiv2026

ACDC: Adaptive Curriculum Planning with Dynamic Contrastive Control for Goal-Conditioned Reinforcement Learning in Robotic Manipulation

Goal-conditioned reinforcement learning has shown considerable potential in robotic manipulation; however, existing approaches remain limited by their reliance on prioritizing collected experience, resulting in suboptimal performance across diverse tasks. Inspired by human learning behaviors, we propose a more comprehensive learning paradigm, ACDC, which integrates multidimensional Adaptive Curriculum (AC) Planning with Dynamic Contrastive (DC) Control to guide the agent along a well-designed learning trajectory. More specifically, at the planning level, the AC component schedules the learning curriculum by dynamically balancing diversity-driven exploration and quality-driven exploitation based on the agent's success rate and training progress. At the control level, the DC component implements the curriculum plan through norm-constrained contrastive learning, enabling magnitude-guided experience selection aligned with the current curriculum focus. Extensive experiments on challenging robotic manipulation tasks demonstrate that ACDC consistently outperforms the state-of-the-art baselines in both sample efficiency and final task success rate.

cs.RO↗

arXiv2021

Sim2Sim Evaluation of a Novel Data-Efficient Differentiable Physics Engine for Tensegrity Robots

Learning policies in simulation is promising for reducing human effort when training robot controllers. This is especially true for soft robots that are more adaptive and safe but also more difficult to accurately model and control. The sim2real gap is the main barrier to successfully transfer policies from simulation to a real robot. System identification can be applied to reduce this gap but traditional identification methods require a lot of manual tuning. Data-driven alternatives can tune dynamical models directly from data but are often data hungry, which also incorporates human effort in collecting data. This work proposes a data-driven, end-to-end differentiable simulator focused on the exciting but challenging domain of tensegrity robots. To the best of the authors' knowledge, this is the first differentiable physics engine for tensegrity robots that supports cable, contact, and actuation modeling. The aim is to develop a reasonably simplified, data-driven simulation, which can learn approximate dynamics with limited ground truth data. The dynamics must be accurate enough to generate policies that can be transferred back to the ground-truth system. As a first step in this direction, the current work demonstrates sim2sim transfer, where the unknown physical model of MuJoCo acts as a ground truth system. Two different tensegrity robots are used for evaluation and learning of locomotion policies, a 6-bar and a 3-bar tensegrity. The results indicate that only 0.25\% of ground truth data are needed to train a policy that works on the ground truth system when the differentiable engine is used for training against training the policy directly on the ground truth system.

cs.RO↗

arXiv2023

Accelerating Reinforcement Learning of Robotic Manipulations via Feedback from Large Language Models

Reinforcement Learning (RL) plays an important role in the robotic manipulation domain since it allows self-learning from trial-and-error interactions with the environment. Still, sample efficiency and reward specification seriously limit its potential. One possible solution involves learning from expert guidance. However, obtaining a human expert is impractical due to the high cost of supervising an RL agent, and developing an automatic supervisor is a challenging endeavor. Large Language Models (LLMs) demonstrate remarkable abilities to provide human-like feedback on user inputs in natural language. Nevertheless, they are not designed to directly control low-level robotic motions, as their pretraining is based on vast internet data rather than specific robotics data. In this paper, we introduce the Lafite-RL (Language agent feedback interactive Reinforcement Learning) framework, which enables RL agents to learn robotic tasks efficiently by taking advantage of LLMs' timely feedback. Our experiments conducted on RLBench tasks illustrate that, with simple prompt design in natural language, the Lafite-RL agent exhibits improved learning capabilities when guided by an LLM. It outperforms the baseline in terms of both learning efficiency and success rate, underscoring the efficacy of the rewards provided by an LLM.

cs.RO↗

arXiv2020

Robot-Supervised Learning for Object Segmentation

To be effective in unstructured and changing environments, robots must learn to recognize new objects. Deep learning has enabled rapid progress for object detection and segmentation in computer vision; however, this progress comes at the price of human annotators labeling many training examples. This paper addresses the problem of extending learning-based segmentation methods to robotics applications where annotated training data is not available. Our method enables pixelwise segmentation of grasped objects. We factor the problem of segmenting the object from the background into two sub-problems: (1) segmenting the robot manipulator and object from the background and (2) segmenting the object from the manipulator. We propose a kinematics-based foreground segmentation technique to solve (1). To solve (2), we train a self-recognition network that segments the robot manipulator. We train this network without human supervision, leveraging our foreground segmentation technique from (1) to label a training set of images containing the robot manipulator without a grasped object. We demonstrate experimentally that our method outperforms state-of-the-art adaptable in-hand object segmentation. We also show that a training set composed of automatically labelled images of grasped objects improves segmentation performance on a test set of images of the same objects in the environment.

cs.RO↗

arXiv2016

Collective Robot Reinforcement Learning with Distributed Asynchronous Guided Policy Search

In principle, reinforcement learning and policy search methods can enable robots to learn highly complex and general skills that may allow them to function amid the complexity and diversity of the real world. However, training a policy that generalizes well across a wide range of real-world conditions requires far greater quantity and diversity of experience than is practical to collect with a single robot. Fortunately, it is possible for multiple robots to share their experience with one another, and thereby, learn a policy collectively. In this work, we explore distributed and asynchronous policy learning as a means to achieve generalization and improved training times on challenging, real-world manipulation tasks. We propose a distributed and asynchronous version of Guided Policy Search and use it to demonstrate collective policy learning on a vision-based door opening task using four robots. We show that it achieves better generalization, utilization, and training times than the single robot alternative.

cs.LG↗

arXiv2025

SimLauncher: Launching Sample-Efficient Real-world Robotic Reinforcement Learning via Simulation Pre-training

Autonomous learning of dexterous, long-horizon robotic skills has been a longstanding pursuit of embodied AI. Recent advances in robotic reinforcement learning (RL) have demonstrated remarkable performance and robustness in real-world visuomotor control tasks. However, applying RL in the real world faces challenges such as low sample efficiency, slow exploration, and significant reliance on human intervention. In contrast, simulators offer a safe and efficient environment for extensive exploration and data collection, while the visual sim-to-real gap, often a limiting factor, can be mitigated using real-to-sim techniques. Building on these, we propose SimLauncher, a novel framework that combines the strengths of real-world RL and real-to-sim-to-real approaches to overcome these challenges. Specifically, we first pre-train a visuomotor policy in the digital twin simulation environment, which then benefits real-world RL in two ways: (1) bootstrapping target values using extensive simulated demonstrations and real-world demonstrations derived from pre-trained policy rollouts, and (2) Incorporating action proposals from the pre-trained policy for better exploration. We conduct comprehensive experiments across multi-stage, contact-rich, and dexterous hand manipulation tasks. Compared to prior real-world RL approaches, SimLauncher significantly improves sample efficiency and achieves near-perfect success rates. We hope this work serves as a proof of concept and inspires further research on leveraging large-scale simulation pre-training to benefit real-world robotic RL.

cs.RO↗

arXiv2020

A Generalized Robotic Handwriting Learning System based on Dynamic Movement Primitives (DMPs)

Learning from demonstration (LfD) is a powerful learning method to enable a robot to infer how to perform a task given one or more human demonstrations of the desired task. By learning from end-user demonstration rather than requiring that a domain expert manually programming each skill, robots can more readily be applied to a wider range of real-world applications. Writing robots, as one application of LfD, has become a challenging research topic due to the complexity of human handwriting trajectories. In this paper, we introduce a generalized handwriting-learning system for a physical robot to learn from examples of humans' handwriting to draw alphanumeric characters. Our robotic system is able to rewrite letters imitating the way human demonstrators write and create new letters in a similar writing style. For this system, we develop an augmented dynamic movement primitive (DMP) algorithm, DMP*, which strengthens the robustness and generalization ability of our robotic system.

cs.RO↗

arXiv2020

Plan-Space State Embeddings for Improved Reinforcement Learning

Robot control problems are often structured with a policy function that maps state values into control values, but in many dynamic problems the observed state can have a difficult to characterize relationship with useful policy actions. In this paper we present a new method for learning state embeddings from plans or other forms of demonstrations such that the embedding space has a specified geometric relationship with the demonstrations. We present a novel variational framework for learning these embeddings that attempts to optimize trajectory linearity in the learned embedding space. We show how these embedding spaces can then be used as an augmentation to the robot state in reinforcement learning problems. We use kinodynamic planning to generate training trajectories for some example environments, and then train embedding spaces for these environments. We show empirically that observing a system in the learned embedding space improves the performance of policy gradient reinforcement learning algorithms, particularly by reducing the variance between training runs. Our technique is limited to environments where demonstration data is available, but places no limits on how that data is collected. Our embedding technique provides a way to transfer domain knowledge from existing technologies such as planning and control algorithms, into more flexible policy learning algorithms, by creating an abstract representation of the robot state with meaningful geometry.

cs.LG↗

arXiv2024

Video2Reward: Generating Reward Function from Videos for Legged Robot Behavior Learning

Learning behavior in legged robots presents a significant challenge due to its inherent instability and complex constraints. Recent research has proposed the use of a large language model (LLM) to generate reward functions in reinforcement learning, thereby replacing the need for manually designed rewards by experts. However, this approach, which relies on textual descriptions to define learning objectives, fails to achieve controllable and precise behavior learning with clear directionality. In this paper, we introduce a new video2reward method, which directly generates reward functions from videos depicting the behaviors to be mimicked and learned. Specifically, we first process videos containing the target behaviors, converting the motion information of individuals in the videos into keypoint trajectories represented as coordinates through a video2text transforming module. These trajectories are then fed into an LLM to generate the reward function, which in turn is used to train the policy. To enhance the quality of the reward function, we develop a video-assisted iterative reward refinement scheme that visually assesses the learned behaviors and provides textual feedback to the LLM. This feedback guides the LLM to continually refine the reward function, ultimately facilitating more efficient behavior learning. Experimental results on tasks involving bipedal and quadrupedal robot motion control demonstrate that our method surpasses the performance of state-of-the-art LLM-based reward generation methods by over 37.6% in terms of human normalized score. More importantly, by switching video inputs, we find our method can rapidly learn diverse motion behaviors such as walking and running.

cs.RO↗

arXiv2026

R2D-RL: A RoboCup 2D Soccer Environment for Multi-Agent Reinforcement Learning

Robot soccer is a challenging testbed for multi-agent reinforcement learning because it combines partial observability, cooperative and adversarial interaction, sparse rewards, and long-horizon tactical behavior. RoboCup 2D Soccer Simulation (RCSS2D) provides a mature robot-soccer platform, but its competition-oriented server-client architecture is difficult to use directly with modern Python-based MARL workflows. We introduce R2D-RL, a reinforcement learning environment that connects RCSS2D and HELIOS-based player clients to a Python MARL interface through shared-memory communication and cycle-level synchronization. R2D-RL supports full-field and scenario-based training with configurable opponents, Base discrete and Hybrid parameterized action spaces, action masks, expected possession value (EPV)-based reward shaping, and parallel execution. We provide front-goal scenarios and an 11-vs-11 full-field benchmark, together with baseline results.

cs.AI↗

arXiv2026

DataMIL: Selecting Data for Robot Imitation Learning with Datamodels

Recently, the robotics community has amassed ever larger and more diverse datasets to train generalist policies. However, while these policies achieve strong mean performance across a variety of tasks, they often underperform on individual, specialized tasks and require further tuning on newly acquired task-specific data. Combining task-specific data with carefully curated subsets of large prior datasets via co-training can produce better specialized policies, but selecting data naively may actually harm downstream performance. To address this, we introduce DataMIL, a data selection framework built on the datamodels paradigm that reasons about data selection in an end-to-end manner, using the policy itself to identify which data points will most improve performance. Unlike standard practices that filter data using human notions of quality (e.g., based on semantic or visual similarity), DataMIL directly optimizes data selection for task success, allowing us to select data that improves the policy while dropping data that degrade it. To avoid performing expensive rollouts in the environment during selection, we introduce a surrogate loss function on task-specific data, allowing us to use DataMIL in the real world without degrading performance. We validate our approach on 60+ simulation and real-world manipulation tasks, notably showing successful data selection from the largest open collections of robot datasets (OXE); demonstrating consistent gains in success rates over prior works. Our results underscore the importance of end-to-end, performance-aware data selection for unlocking the potential of large prior datasets in robotics. More information at https://robin-lab.cs.utexas.edu/datamodels4imitation/

cs.RO↗