One-Step Model Predictive Path Integral for Manipulator Motion Planning Using Configuration Space Distance Fields

2025-08-31

AI Summary

Traditional optimization methods for high-dimensional robotic arm motion planning suffer from local minima and失效 SDF gradient information, while sampling-based approaches like MPPI incur high computational overhead and require careful, often non-intuitive, cost function design. To address these issues, this paper proposes a single-step receding-horizon planning framework that integrates Configuration-space Distance Fields (CDF) with Model Predictive Path Integral (MPPI) control. We introduce CDF into MPPI for the first time, leveraging its robust gradient for direct joint-space navigation. A physically consistent configuration-space cost function is uniformly formulated, and the prediction horizon is reduced to one step—dramatically lowering collision-checking and sampling complexity. Evaluated in 2D environments and 7-DOF Franka Emika simulations, our method achieves near 100% planning success rate, real-time control frequency exceeding 750 Hz, and significantly outperforms conventional optimization and standard MPPI in both obstacle avoidance and computational efficiency.

Technology Category

• Intelligent Robots: Motion and Path Planning
• Planning, Routing, and Scheduling: Replanning and Plan Repair
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Motion planning for robotic manipulators is a fundamental problem in robotics. Classical optimization-based methods typically rely on the gradients of signed distance fields (SDFs) to impose collision-avoidance constraints. However, these methods are susceptible to local minima and may fail when the SDF gradients vanish. Recently, Configuration Space Distance Fields (CDFs) have been introduced, which directly model distances in the robot's configuration space. Unlike workspace SDFs, CDFs are differentiable almost everywhere and thus provide reliable gradient information. On the other hand, gradient-free approaches such as Model Predictive Path Integral (MPPI) control leverage long-horizon rollouts to achieve collision avoidance. While effective, these methods are computationally expensive due to the large number of trajectory samples, repeated collision checks, and the difficulty of designing cost functions with heterogeneous physical units. In this paper, we propose a framework that integrates CDFs with MPPI to enable direct navigation in the robot's configuration space. Leveraging CDF gradients, we unify the MPPI cost in joint-space and reduce the horizon to one step, substantially cutting computation while preserving collision avoidance in practice. We demonstrate that our approach achieves nearly 100% success rates in 2D environments and consistently high success rates in challenging 7-DOF Franka manipulator simulations with complex obstacles. Furthermore, our method attains control frequencies exceeding 750 Hz, substantially outperforming both optimization-based and standard MPPI baselines. These results highlight the effectiveness and efficiency of the proposed CDF-MPPI framework for high-dimensional motion planning.

Problem

• Integrating CDFs with MPPI for efficient manipulator motion planning
• Addressing local minima and gradient vanishing in collision avoidance
• Achieving real-time control in high-dimensional configuration spaces

Innovation

• Combines CDFs with MPPI for configuration space navigation
• Uses CDF gradients to unify joint-space MPPI cost
• Reduces horizon to one step for high-frequency control