Local Reactive Control for Mobile Manipulators with Whole-Body Safety in Complex Environments

2025-01-06

AI Summary

Real-time whole-body safety control of mobile manipulators in narrow, cluttered environments remains challenging—conventional reactive methods struggle to balance computational efficiency and holistic safety under high-dimensional state spaces and strongly coupled constraints. Method: This paper proposes a multi-step optimization framework grounded in spatial propagation along kinematic chains. It extends traditional single-step temporal optimization into a spatial-domain multi-step paradigm, integrating link-level decoupled constraint modeling and geometrically exact per-chain collision avoidance. The approach employs Augmented Lagrangian Differential Dynamic Programming (AL-DDP), forward propagation over serial kinematic chains, and free-space-driven geometric collision detection. Contribution/Results: Experiments demonstrate significant improvements in safety, motion efficiency, and task success rate. Crucially, the method maintains robust performance even in scenarios where conventional approaches fail, validating its effectiveness for complex, constrained operational environments.

Technology Category

• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Search and Optimization: Mixed Discrete/Continuous Search
• Intelligent Robots: Motion and Path Planning

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web
• Security and Privacy: Large-scale security measurements

Abstract

Mobile manipulators typically encounter significant challenges in navigating narrow, cluttered environments due to their high-dimensional state spaces and complex kinematics. While reactive methods excel in dynamic settings, they struggle to efficiently incorporate complex, coupled constraints across the entire state space. In this work, we present a novel local reactive controller that reformulates the time-domain single-step problem into a multi-step optimization problem in the spatial domain, leveraging the propagation of a serial kinematic chain. This transformation facilitates the formulation of customized, decoupled link-specific constraints, which is further solved efficiently with augmented Lagrangian differential dynamic programming (AL-DDP). Our approach naturally absorbs spatial kinematic propagation in the forward pass and processes all link-specific constraints simultaneously during the backward pass, enhancing both constraint management and computational efficiency. Notably, in this framework, we formulate collision avoidance constraints for each link using accurate geometric models with extracted free regions, and this improves the maneuverability of the mobile manipulator in narrow, cluttered spaces. Experimental results showcase significant improvements in safety, efficiency, and task completion rates. These findings underscore the robustness of the proposed method, particularly in narrow, cluttered environments where conventional approaches could falter. The open-source project can be found at https://github.com/Chunx1nZHENG/MM-with-Whole-Body-Safety-Release.git.

Problem

• Mobile Robotics
• Real-time Control
• Complex Environments

Innovation

• Multi-step Optimization
• AL-DDP Algorithm
• Complex Environment Adaptability