Estimated Informed Anytime Search for Sampling-Based Planning via Adaptive Sampler

2025-08-29

AI Summary

Sampling-based path planning in high-dimensional continuous spaces suffers from slow initial convergence and necessitates full-space resampling when no solution exists. To address these challenges, this paper proposes Multi-Informed Trees (MIT*). Methodologically, MIT* introduces a novel prior-cost-driven heuristic ensemble that operates without requiring an initial feasible solution, enabling efficient early-stage guidance. It further integrates adaptive sampling, sparse collision checking, and a path-length-dependent delayed backtracking search. Evaluated in configuration spaces ℝ⁴–ℝ¹⁶, MIT* achieves significantly faster convergence and higher solution success rates compared to state-of-the-art single-query sampling planners. Experimental results on both simulated and real-robot tasks demonstrate MIT*’s superior computational efficiency, improved path quality (e.g., shorter, smoother trajectories), and enhanced robustness in complex, cluttered environments.

Technology Category

• Search and Optimization: Sampling/Simulation-based Search
• Intelligent Robots: Motion and Path Planning
• Planning, Routing, and Scheduling: Replanning and Plan Repair

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Path planning in robotics often involves solving continuously valued, high-dimensional problems. Popular informed approaches include graph-based searches, such as A*, and sampling-based methods, such as Informed RRT*, which utilize informed set and anytime strategies to expedite path optimization incrementally. Informed sampling-based planners define informed sets as subsets of the problem domain based on the current best solution cost. However, when no solution is found, these planners re-sample and explore the entire configuration space, which is time-consuming and computationally expensive. This article introduces Multi-Informed Trees (MIT*), a novel planner that constructs estimated informed sets based on prior admissible solution costs before finding the initial solution, thereby accelerating the initial convergence rate. Moreover, MIT* employs an adaptive sampler that dynamically adjusts the sampling strategy based on the exploration process. Furthermore, MIT* utilizes length-related adaptive sparse collision checks to guide lazy reverse search. These features enhance path cost efficiency and computation times while ensuring high success rates in confined scenarios. Through a series of simulations and real-world experiments, it is confirmed that MIT* outperforms existing single-query, sampling-based planners for problems in R^4 to R^16 and has been successfully applied to real-world robot manipulation tasks. A video showcasing our experimental results is available at: https://youtu.be/30RsBIdexTU

Problem

• Accelerates initial solution convergence in high-dimensional path planning
• Reduces time-consuming exploration of entire configuration space
• Enhances path cost efficiency and computation times adaptively

Innovation

• Estimated informed sets accelerate initial convergence
• Adaptive sampler dynamically adjusts sampling strategy
• Length-related adaptive sparse collision checks guide search