GeoPF: Infusing Geometry into Potential Fields for Reactive Planning in Non-trivial Environments

TL;DR

GeoPF introduces a geometric-aware potential field framework that improves reactive robot motion planning by explicitly modeling environmental primitives, leading to higher success rates and lower computational costs in complex, dynamic environments.

Contribution

The paper presents GeoPF, a novel reactive planning method that incorporates geometric primitives into potential fields, enhancing accuracy and efficiency over traditional point-based approaches.

Findings

Higher success rates in obstacle avoidance

Reduced tuning complexity with a single parameter set

Significantly lower computational costs (up to 100x)

Abstract

Reactive intelligence remains one of the cornerstones of versatile robotics operating in cluttered, dynamic, and human-centred environments. Among reactive approaches, potential fields (PF) continue to be widely adopted due to their simplicity and real-time applicability. However, existing PF methods typically oversimplify environmental representations by relying on isotropic, point- or sphere-based obstacle approximations. In human-centred settings, this simplification results in overly conservative paths, cumbersome tuning, and computational overhead -- even breaking real-time requirements. In response, we propose the Geometric Potential Field (GeoPF), a reactive motion-planning framework that explicitly infuses geometric primitives -- points, lines, planes, cubes, and cylinders -- their structure and spatial relationship in modulating the real-time repulsive response. Extensive quantitative analyses consistently show GeoPF's higher success rates, reduced tuning complexity (a single parameter set across experiments), and substantially lower computational costs (up to 2 orders of magnitude) compared to traditional PF methods. Real-world experiments further validate GeoPF reliability, robustness, and practical ease of deployment, as well as its scalability to whole-body avoidance. GeoPF provides a fresh perspective on reactive planning problems driving geometric-aware temporal motion generation, enabling flexible and low-latency motion planning suitable for modern robotic applications.