DiPPeR: Diffusion-based 2D Path Planner applied on Legged Robots

TL;DR

DiPPeR introduces a diffusion-based 2D path planning framework for quadrupedal robots, enabling faster and consistent trajectory generation in complex environments, validated on real robots and maze scenarios.

Contribution

The paper presents a novel diffusion-driven path planning method, including a dataset generator and CNN-based pipeline, specifically designed for quadrupedal robot navigation.

Findings

23x faster trajectory generation compared to traditional methods

87% success rate in producing feasible paths across various maps

Effective deployment on real quadrupedal robots in real-world scenarios

Abstract

In this work, we present DiPPeR, a novel and fast 2D path planning framework for quadrupedal locomotion, leveraging diffusion-driven techniques. Our contributions include a scalable dataset generator for map images and corresponding trajectories, an image-conditioned diffusion planner for mobile robots, and a training/inference pipeline employing CNNs. We validate our approach in several mazes, as well as in real-world deployment scenarios on Boston Dynamic's Spot and Unitree's Go1 robots. DiPPeR performs on average 23 times faster for trajectory generation against both search based and data driven path planning algorithms with an average of 87% consistency in producing feasible paths of various length in maps of variable size, and obstacle structure. Website: https://rpl-cs-ucl.github.io/DiPPeR