AutoDeep: Automatic Design of Safe, High- Performance and Compact Deep Learning Models for Autonomous Vehicles
Deep Neural Networks (DNN) are increasingly being used to support decision-making in autonomous vehicles. In this project we develop a framework called AutoDeep to achieve performance, compactness, and robustness in design and customization of DNN for intention detection and behavior prediction as two safety-critical applications in road and construction autonomous vehicles.
Concluded
Start
2020-01-01
Conclusion
2024-01-02
Collaboration partners
Project manager at MDU
Deep Neural Networks (DNN) are increasingly being used to support decision-making in autonomous vehicles. In this project we develop a framework called AutoDeep to achieve performance, compactness, and robustness in design and customization of DNN for intention detection and behavior prediction as two safety-critical applications in road and construction autonomous vehicles. To the best of our knowledge, it would be the first framework for designing DNNs that considers performance, compactness, and robustness as well.
Project objectives
To provide new techniques & tools to produce robust, compact and accurate deep learning models for safe-critical applications in autonomous vehicles.
