Evolutionary Hyperparameter Optimization to Find Lightweight CNN Models for Autonomous Steering
2026-06-29 • Neural and Evolutionary Computing
Neural and Evolutionary ComputingComputer Vision and Pattern RecognitionRobotics
AI summaryⓘ
The authors studied how to make small and fast neural networks that can steer a car by predicting the steering angle from camera images. They used an evolutionary algorithm called (N+M) Evolution Strategy to automatically adjust the network’s design for better performance. Their data came from a small set of driving scenarios collected with an autonomous driving platform. By optimizing a baseline CNN, they made the model much smaller without losing much accuracy. This shows it’s possible to build efficient steering models that work well in real time for self-driving cars.
Convolutional Neural NetworksDense Neural NetworksEvolution StrategyHyperparameter TuningAutonomous DrivingSteering Angle PredictionReal-time SystemsLightweight Models1/5th Success RuleLTU ACTor Platform
Authors
Devson Butani, Ryan Kaddis, Chan-Jin Chung
Abstract
This research investigates the optimization of Convolutional and Dense Neural Networks (CNNs and DNNs) for autonomous steering using the (N+M) Evolution Strategy (ES) with the 1/5th success rule. The primary objective is to develop a lightweight CNN based model capable of real-time steering angle prediction, mimicking human driving behavior on predefined paths. The ES algorithm automates hyperparameter tuning, dynamically adjusting parameters such as filter sizes and layer configurations. Data collection encompasses driving scenarios recorded via the LTU ACTor autonomous driving platform, including variations in path direction and driving style. The very small dataset consists of timestamped images labeled with steering angles and pre-processed to focus on relevant visual information. Initial experiments involve training a baseline CNN model, which is then refined using ES to significantly reduce the size of the model while maintaining competitive predictive accuracy. The results highlight the viability of lightweight neural network architectures for real-time autonomous systems, striking a balance between computational efficiency and performance. This study not only advances research initiatives on the use of evolutionary algorithms for autonomous driving applications but also lays the foundation for the deployment of cost-effective and scalable solutions in self-driving technology.