Neuro-Cognitive Reward Modeling for Human-Centered Autonomous Vehicle Control

TL;DR

This paper introduces an EEG-guided reinforcement learning framework that leverages human cognitive signals to improve autonomous vehicle decision-making, especially in collision avoidance, without interrupting human behavior.

Contribution

It presents a novel neuro-cognitive reward modeling approach using EEG data to enhance RL-based autonomous driving systems, reducing reliance on manual preference data collection.

Findings

EEG signals can predict cognitive responses to environmental changes.

Integrating EEG-based ERP into RL improves collision avoidance.

The framework demonstrates potential for neuro-cognitive feedback in autonomous driving.

Abstract

Recent advancements in computer vision have accelerated the development of autonomous driving. Despite these advancements, training machines to drive in a way that aligns with human expectations remains a significant challenge. Human factors are still essential, as humans possess a sophisticated cognitive system capable of rapidly interpreting scene information and making accurate decisions. Aligning machine with human intent has been explored with Reinforcement Learning with Human Feedback (RLHF). Conventional RLHF methods rely on collecting human preference data by manually ranking generated outputs, which is time-consuming and indirect. In this work, we propose an electroencephalography (EEG)-guided decision-making framework to incorporate human cognitive insights without behaviour response interruption into reinforcement learning (RL) for autonomous driving. We collected EEG signals from 20 participants in a realistic driving simulator and analyzed event-related potentials (ERP) in response to sudden environmental changes. Our proposed framework employs a neural network to predict the strength of ERP based on the cognitive information from visual scene information. Moreover, we explore the integration of such cognitive information into the reward signal of the RL algorithm. Experimental results show that our framework can improve the collision avoidance ability of the RL algorithm, highlighting the potential of neuro-cognitive feedback in enhancing autonomous driving systems. Our project page is: https://alex95gogo.github.io/Cognitive-Reward/.