TruPercept: Trust Modelling for Autonomous Vehicle Cooperative Perception from Synthetic Data

TL;DR

TruPercept introduces a trust-based communication framework for autonomous vehicle perception, utilizing a synthetic dataset to evaluate cooperative perception robustness against unreliable and malicious data sources.

Contribution

The paper presents a novel trust modeling approach for AV communication, along with a synthetic dataset for testing cooperative perception under various trust scenarios.

Findings

Trust modeling improves perception accuracy with unreliable data

Synthetic dataset enables testing of malicious behavior scenarios

Framework supports modular evaluation and ablation studies

Abstract

Inter-vehicle communication for autonomous vehicles (AVs) stands to provide significant benefits in terms of perception robustness. We propose a novel approach for AVs to communicate perceptual observations, tempered by trust modelling of peers providing reports. Based on the accuracy of reported object detections as verified locally, communicated messages can be fused to augment perception performance beyond line of sight and at great distance from the ego vehicle. Also presented is a new synthetic dataset which can be used to test cooperative perception. The TruPercept dataset includes unreliable and malicious behaviour scenarios to experiment with some challenges cooperative perception introduces. The TruPercept runtime and evaluation framework allows modular component replacement to facilitate ablation studies as well as the creation of new trust scenarios we are able to show.