GeoPF: Infusing Geometry into Potential Fields for Reactive Planning in Non-trivial Environments
Yuhe Gong, Riddhiman Laha, Luis Figueredo
AI summary
Problem
Traditional potential field methods oversimplify obstacles using isotropic point or sphere approximations, causing overly conservative paths, cumbersome tuning, and computational overhead that hinder real-time reactivity in dynamic, human-centered environments.
Approach
GeoPF explicitly models obstacles as geometric primitives like lines, planes, cubes, and cylinders, using their structural properties and spatial relationships to directly modulate real-time repulsive forces.
Key results
- Higher success rates in cluttered and dynamic settings
- Single universal parameter set eliminating complex tuning
- Up to two orders of magnitude reduction in computational cost
- Analytical closed-form repulsive forces for common geometric primitives
Why it matters
Enables reliable, low-latency reactive navigation for modern robots in unstructured human-centered spaces without heavy tuning or computational overhead.
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’s reliability, robustness, and practical ease of deployment. 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.