APPLD: Adaptive Planner Parameter Learning from Demonstration

TL;DR

APPLD enables robots to adapt their navigation parameters to new environments through human demonstrations, reducing the need for expert re-tuning and improving navigation performance.

Contribution

APPLD introduces a novel learning framework that adapts existing navigation systems to new environments using demonstration data from teleoperation.

Findings

APPLD outperforms default and expert-tuned parameters.

APPLD surpasses human demonstrator performance.

Verified on two different robots and environments.

Abstract

Existing autonomous robot navigation systems allow robots to move from one point to another in a collision-free manner. However, when facing new environments, these systems generally require re-tuning by expert roboticists with a good understanding of the inner workings of the navigation system. In contrast, even users who are unversed in the details of robot navigation algorithms can generate desirable navigation behavior in new environments via teleoperation. In this paper, we introduce APPLD, Adaptive Planner Parameter Learning from Demonstration, that allows existing navigation systems to be successfully applied to new complex environments, given only a human teleoperated demonstration of desirable navigation. APPLD is verified on two robots running different navigation systems in different environments. Experimental results show that APPLD can outperform navigation systems with the default and expert-tuned parameters, and even the human demonstrator themselves.