A part of Autonomous car powered by "Deep Reinforcement Learning Model" and the engine supported by YoLoP. The model trained in the simulated environment called Carla and tested in the real-time.
AI model dynamically quantized for INT8 and deployed onto NVIDIA's Jetson Nano version. Eventhough the visual preception subsystem can work on resource-constrained system, yet the DQL model's performance is not up to the level. Feel free to check out the demo for entire autonomous system.
The project is a part of undergrad program of Anna University, Chennai, India, and all rights reserved to the university
@article{bhavanisankari2023ai,
title={Design and Training of AI Agent using Deep Q - Learning and Carla},
author={Bhavanisankari, S. and William, V. and Sachin, A.},
journal={International Journal of Scientific Research in Science, Engineering and Technology (IJSRSET)},
pages={74-81},
year={2023}
}