Reactive Neural Path Planning with Dynamic Obstacle Avoidance in a Condensed Configuration Space

TL;DR

This paper introduces a biologically inspired neural network-based path planning method that enables real-time obstacle avoidance in dynamic environments by operating in a condensed configuration space.

Contribution

It presents a novel approach combining self-organizing neural networks and cognitive maps for fast, reactive path planning in dynamic settings, validated on real robots and simulations.

Findings

Re-planning within 0.02 seconds demonstrates real-time capability.

The method effectively handles dynamic obstacle avoidance.

Compared favorably to sample-based planners in experiments.

Abstract

We present a biologically inspired approach for path planning with dynamic obstacle avoidance. Path planning is performed in a condensed configuration space of a robot generated by self-organizing neural networks (SONN). The robot itself and static as well as dynamic obstacles are mapped from the Cartesian task space into the configuration space by precomputed kinematics. The condensed space represents a cognitive map of the environment, which is inspired by place cells and the concept of cognitive maps in mammalian brains. The generation of training data as well as the evaluation were performed on a real industrial robot accompanied by simulations. To evaluate the reactive collision-free online planning within a changing environment, a demonstrator was realized. Then, a comparative study regarding sample-based planners was carried out. So we could show that the robot is able to operate in dynamically changing environments and re-plan its motion trajectories within impressing 0.02 seconds, which proofs the real-time capability of our concept.