Transport and Smart Mobility
Drones, Unmanned Aerial Vehicles (UAV) and Unmanned Underwater Vehicles (UUV) are used in many domains like automotive surveillance, border surveillance or acoustic signal classification. These applications need to embed neural networks with ultra-low power consumption which is one the most important parameters to evaluate. Then, the goal will be to evaluate the performance and the power consumption on the ANN and SNN accelerators developed in ANDANTE and to compare them with the other existing accelerators. In this application domain, the project partners will evaluate and demonstrate in the Use cases 3.1 Drones/USV, Use case 3.2 Underwater Acoustic Signal Classification, and Use case 3.4 Robust Autonomous Landing what neuromorphic hardware type can add to such kind of transport and smart mobility applications.
Furthermore, self-driving and electrically powered cars are two trends that shaped the automotive market and research in the last years. Neuromorphic hardware will greatly support efficient implementations of both trends due to their potential for higher performance, lower cost and energy consumption in comparison to state-of-the-arts solutions for data processing in cars. Hence, the partners will evaluate and demonstrate the potential of neuromorphic circuits with applications like Lidar 3D traffic object classification that are important components of a self-driving car system. Within this domain, the self-driving applications will be demonstrated in the Use case 3.3 3D Object Detection and Classification of Road Users based on LiDAR and camera and Use case 3.5 Path planning for autonomous steering.