Deep Reactive Policy: Learning Reactive Manipulator Motion Planning for Dynamic Environments

• URL: http://arxiv.org/abs/2509.06953v1
• Date: Mon, 08 Sep 2025 17:59:35 GMT
• Title: Deep Reactive Policy: Learning Reactive Manipulator Motion Planning for Dynamic Environments
• Authors: Jiahui Yang, Jason Jingzhou Liu, Yulong Li, Youssef Khaky, Kenneth Shaw, Deepak Pathak,
• Abstract summary: Deep Reactive Policy is a visuo-motor neural motion policy designed for reactive motion generation in diverse dynamic environments.<n>At its core is IMPACT, a transformer-based neural motion policy pretrained on 10 million generated expert trajectories.<n>We enhance IMPACT's dynamic obstacle avoidance at inference time using DCP-RMP, a locally reactive goal-proposal module.
• Score: 35.192151100313836
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Generating collision-free motion in dynamic, partially observable environments is a fundamental challenge for robotic manipulators. Classical motion planners can compute globally optimal trajectories but require full environment knowledge and are typically too slow for dynamic scenes. Neural motion policies offer a promising alternative by operating in closed-loop directly on raw sensory inputs but often struggle to generalize in complex or dynamic settings. We propose Deep Reactive Policy (DRP), a visuo-motor neural motion policy designed for reactive motion generation in diverse dynamic environments, operating directly on point cloud sensory input. At its core is IMPACT, a transformer-based neural motion policy pretrained on 10 million generated expert trajectories across diverse simulation scenarios. We further improve IMPACT's static obstacle avoidance through iterative student-teacher finetuning. We additionally enhance the policy's dynamic obstacle avoidance at inference time using DCP-RMP, a locally reactive goal-proposal module. We evaluate DRP on challenging tasks featuring cluttered scenes, dynamic moving obstacles, and goal obstructions. DRP achieves strong generalization, outperforming prior classical and neural methods in success rate across both simulated and real-world settings. Video results and code available at https://deep-reactive-policy.com

Related papers

• Closed-Loop Action Chunks with Dynamic Corrections for Training-Free Diffusion Policy [52.106797722292896]
  We present DCDP, a Dynamic Closed-Loop Diffusion Policy framework that integrates chunk-based action generation with real-time correction.<n>In dynamic PushT simulations, DCDP improves adaptability by 19% without retraining while requiring only 5% additional computation.
  arXiv  Detail & Related papers  (2026-03-02T15:04:18Z)
• Towards Learning Scalable Agile Dynamic Motion Planning for Robosoccer Teams with Policy Optimization [0.0]
  Dynamic Motion Planning for Multi-Agent Systems in the presence of obstacles is a universal and unsolved problem.<n>We present a learning-based dynamic navigation model and show our model working on a simple environment in the concept of a simple Robosoccer Game.
  arXiv  Detail & Related papers  (2025-02-08T11:13:07Z)
• Dynamic Neural Potential Field: Online Trajectory Optimization in Presence of Moving Obstacles [40.8414230686474]
  We address a task of local trajectory planning for the mobile robot in the presence of static and dynamic obstacles. We develop an approach, where repulsive potential is estimated by the neural model. We deploy our approach on Husky UGV mobile platform, which move through the office corridors under proposed MPC local trajectory planner.
  arXiv  Detail & Related papers  (2024-10-09T12:27:09Z)
• FLD: Fourier Latent Dynamics for Structured Motion Representation and Learning [19.491968038335944]
  We introduce a self-supervised, structured representation and generation method that extracts spatial-temporal relationships in periodic or quasi-periodic motions. Our work opens new possibilities for future advancements in general motion representation and learning algorithms.
  arXiv  Detail & Related papers  (2024-02-21T13:59:21Z)
• Latent Exploration for Reinforcement Learning [87.42776741119653]
  In Reinforcement Learning, agents learn policies by exploring and interacting with the environment. We propose LATent TIme-Correlated Exploration (Lattice), a method to inject temporally-correlated noise into the latent state of the policy network.
  arXiv  Detail & Related papers  (2023-05-31T17:40:43Z)
• Learning Robust Policy against Disturbance in Transition Dynamics via State-Conservative Policy Optimization [63.75188254377202]
  Deep reinforcement learning algorithms can perform poorly in real-world tasks due to discrepancy between source and target environments. We propose a novel model-free actor-critic algorithm to learn robust policies without modeling the disturbance in advance. Experiments in several robot control tasks demonstrate that SCPO learns robust policies against the disturbance in transition dynamics.
  arXiv  Detail & Related papers  (2021-12-20T13:13:05Z)
• Nonprehensile Riemannian Motion Predictive Control [57.295751294224765]
  We introduce a novel Real-to-Sim reward analysis technique to reliably imagine and predict the outcome of taking possible actions for a real robotic platform. We produce a closed-loop controller to reactively push objects in a continuous action space. We observe that RMPC is robust in cluttered as well as occluded environments and outperforms the baselines.
  arXiv  Detail & Related papers  (2021-11-15T18:50:04Z)
• Hierarchical Neural Dynamic Policies [50.969565411919376]
  We tackle the problem of generalization to unseen configurations for dynamic tasks in the real world while learning from high-dimensional image input. We use hierarchical deep policy learning framework called Hierarchical Neural Dynamical Policies (H-NDPs) H-NDPs form a curriculum by learning local dynamical system-based policies on small regions in state-space. We show that H-NDPs are easily integrated with both imitation as well as reinforcement learning setups and achieve state-of-the-art results.
  arXiv  Detail & Related papers  (2021-07-12T17:59:58Z)
• Neural Dynamic Policies for End-to-End Sensorimotor Learning [51.24542903398335]
  The current dominant paradigm in sensorimotor control, whether imitation or reinforcement learning, is to train policies directly in raw action spaces. We propose Neural Dynamic Policies (NDPs) that make predictions in trajectory distribution space. NDPs outperform the prior state-of-the-art in terms of either efficiency or performance across several robotic control tasks.
  arXiv  Detail & Related papers  (2020-12-04T18:59:32Z)
• Multi-Task Reinforcement Learning based Mobile Manipulation Control for Dynamic Object Tracking and Grasping [17.2022039806473]
  A multi-task reinforcement learning-based mobile manipulation control framework is proposed to achieve general dynamic object tracking and grasping. Experiments show that our policy trained can adapt to unseen random dynamic trajectories with about 0.1m tracking error and 75% grasping success rate.
  arXiv  Detail & Related papers  (2020-06-07T21:18:36Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.