Autonomous Driving using Spiking Neural Networks on Dynamic Vision Sensor Data: A Case Study of Traffic Light Change Detection

TL;DR

This paper explores the application of spiking neural networks to autonomous driving tasks using dynamic vision sensor data, specifically focusing on traffic light change detection in realistic driving scenarios.

Contribution

It presents a case study demonstrating the feasibility of SNNs for complex autonomous driving tasks in photo-realistic environments, advancing beyond simple simulations.

Findings

SNNs can effectively detect traffic light changes in realistic driving scenes.

The approach shows promise for low-power, real-time autonomous driving applications.

Abstract

Autonomous driving is a challenging task that has gained broad attention from both academia and industry. Current solutions using convolutional neural networks require large amounts of computational resources, leading to high power consumption. Spiking neural networks (SNNs) provide an alternative computational model to process information and make decisions. This biologically plausible model has the advantage of low latency and energy efficiency. Recent work using SNNs for autonomous driving mostly focused on simple tasks like lane keeping in simplified simulation environments. This paper studies SNNs on photo-realistic driving scenes in the CARLA simulator, which is an important step toward using SNNs on real vehicles. The efficacy and generalizability of the method will be investigated.