A Multi-Behavior Planning Framework for Robot Guide

TL;DR

This paper introduces a multi-behavior planning framework for robot guides that combines human-aware navigation with interactive behavior planning using Monte Carlo Tree Search, improving communication and assistance.

Contribution

It presents a novel Monte Carlo Tree Search-based framework that integrates interactive behavior planning and human motion prediction for robot guiding tasks.

Findings

Enhanced guidance accuracy in simulation and real-world tests

Improved human-robot interaction effectiveness

Outperforms existing socially-aware navigation models

Abstract

The guiding task of a mobile robot requires not only human-aware navigation, but also appropriate yet timely interaction for active instruction. State-of-the-art tour-guide models limit their socially-aware consideration to adapting to users' motion, ignoring the interactive behavior planning to fulfill the communicative demands. We propose a multi-behavior planning framework based on Monte Carlo Tree Search to better assist users to understand confusing scene contexts, select proper paths and timely arrive at the destination. To provide proactive guidance, we construct a sampling-based probability model of human motion to consider the interrelated effects between robots and humans. We validate our method both in simulation and real-world experiments along with performance comparison with state-of-the-art models.