SABER: Data-Driven Motion Planner for Autonomously Navigating Heterogeneous Robots

TL;DR

SABER introduces a comprehensive data-driven motion planning framework for heterogeneous robot teams, integrating SMPC, neural networks, distributed Kalman filtering, and deep reinforcement learning to navigate uncertain environments effectively.

Contribution

The paper presents a novel integrated approach combining stochastic control, neural uncertainty estimation, distributed filtering, and deep learning for multi-robot navigation.

Findings

Successfully navigates heterogeneous robots in uncertain environments.

Real-time uncertainty estimation improves obstacle avoidance.

Demonstrated on ground and aerial robots with open-source code.

Abstract

We present an end-to-end online motion planning framework that uses a data-driven approach to navigate a heterogeneous robot team towards a global goal while avoiding obstacles in uncertain environments. First, we use stochastic model predictive control (SMPC) to calculate control inputs that satisfy robot dynamics, and consider uncertainty during obstacle avoidance with chance constraints. Second, recurrent neural networks are used to provide a quick estimate of future state uncertainty considered in the SMPC finite-time horizon solution, which are trained on uncertainty outputs of various simultaneous localization and mapping algorithms. When two or more robots are in communication range, these uncertainties are then updated using a distributed Kalman filtering approach. Lastly, a Deep Q-learning agent is employed to serve as a high-level path planner, providing the SMPC with target positions that move the robots towards a desired global goal. Our complete methods are demonstrated on a ground and aerial robot simultaneously (code available at: https://github.com/AlexS28/SABER).