Growing Trees with an Agent: Accelerating RRTs with Learned, Multi-Step Episodic Exploration

2025/07/09 by Xinyu Wu, Wu, Xinyu
Computer Science · Engineering · #FOS: Computer and information sciences #Human Motion and Animation #Robot Manipulation and Learning #Robotic Path Planning Algorithms #Robotics (cs.RO)

paper · pdf · doi:10.48550/arxiv.2507.06605

openalex publication_date 2025/07/09 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Classical sampling-based motion planners like the RRTs suffer from inefficiencies, particularly in cluttered or high-dimensional spaces, due to their reliance on undirected, random sampling. This paper introduces the Episodic RRT, a novel hybrid planning framework that replaces the primitive of a random point with a learned, multi-step "exploratory episode" generated by a Deep Reinforcement Learning agent. By making the DRL agent the engine of exploration, ERRT transforms the search process from a diffuse, volumetric expansion into a directed, branch-like growth. This paradigm shift yields key advantages: it counters the curse of dimensionality with focused exploration, minimizes expensive collision checks by proactively proposing locally valid paths, and improves connectivity by generating inherently connected path segments. We demonstrate through extensive empirical evaluation across 2D, 3D, and 6D environments that ERRT and its variants consistently and significantly outperform their classical counterparts without any GPU acceleration. In a challenging 6D robotic arm scenario, ERRT achieves a 98% success rate compared to 19% for RRT, is up to 107x faster, reduces collision checks by over 99.6%, and finds initial paths that are nearly 50% shorter. Furthermore, its asymptotically optimal variant, ERRT*, demonstrates vastly superior anytime performance, refining solutions to near-optimality up to 29x faster than standard RRT* in 3D environments. Code: https://xinyuwuu.github.io/Episodic_RRT/.

Citations

• Neural MP: A Generalist Neural Motion Planner
• PathRL: An End-to-End Path Generation Method for Collision Avoidance via Deep Reinforcement Learning
• Neural Informed RRT*: Learning-based Path Planning with Point Cloud State Representations under Admissible Ellipsoidal Constraints
• Learning-based Near-optimal Motion Planning for Intelligent Vehicles with Uncertain Dynamics
• Learning-based Motion Planning in Dynamic Environments Using GNNs and Temporal Encoding
• Mobile Robot Path Planning in Dynamic Environments through Globally Guided Reinforcement Learning
• Learned Critical Probabilistic Roadmaps for Robotic Motion Planning
• Motion Planning Networks: Bridging the Gap Between Learning-based and\n Classical Motion Planners
• Deeply Informed Neural Sampling for Robot Motion Planning
• Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actor
• A New Approach to Time-Optimal Path Parameterization based on Reachability Analysis
• Sampling-based Algorithms for Optimal Motion Planning
• Probabilistic roadmaps for path planning in high-dimensional configuration spaces

Related

• Zonal RL-RRT: Integrated RL-RRT Path Planning with Collision Probability and Zone Connectivity
• Rapidly-exploring Random Forest: Adaptively Exploits Local Structure with Generalised Multi-Trees Motion Planning
• Finding a needle in an exponential haystack: Discrete RRT for exploration of implicit roadmaps in multi-robot motion planning
• AM-RRT*: Informed Sampling-based Planning with Assisting Metric
• Neural Informed RRT*: Learning-based Path Planning with Point Cloud State Representations under Admissible Ellipsoidal Constraints