Context-aware Mamba-based Reinforcement Learning for social robot navigation

TL;DR

This paper introduces CAMRL, a novel context-aware reinforcement learning approach using Mamba, a deep state space model, to improve social robot navigation in pedestrian environments, outperforming existing methods.

Contribution

It presents CAMRL, a new deep-state space model-based reinforcement learning method for social robot navigation, demonstrating superior performance over existing solutions.

Findings

CAMRL achieves higher success rates in navigation tasks.

CAMRL minimizes collisions and maintains safer distances.

The approach outperforms existing methods like CADRL, LSTM-RL, and SARL.

Abstract

Social robot navigation (SRN) is a relevant problem that involves navigating a pedestrian-rich environment in a socially acceptable manner. It is an essential part of making social robots effective in pedestrian-rich settings. The use cases of such robots could vary from companion robots to warehouse robots to autonomous wheelchairs. In recent years, deep reinforcement learning has been increasingly used in research on social robot navigation. Our work introduces CAMRL (Context-Aware Mamba-based Reinforcement Learning). Mamba is a new deep learning-based State Space Model (SSM) that has achieved results comparable to transformers in sequencing tasks. CAMRL uses Mamba to determine the robot's next action, which maximizes the value of the next state predicted by the neural network, enabling the robot to navigate effectively based on the rewards assigned. We evaluate CAMRL alongside existing solutions (CADRL, LSTM-RL, SARL) using a rigorous testing dataset which involves a variety of densities and environment behaviors based on ORCA and SFM, thus, demonstrating that CAMRL achieves higher success rates, minimizes collisions, and maintains safer distances from pedestrians. This work introduces a new SRN planner, showcasing the potential for deep-state space models for robot navigation.