This book offers a comprehensive, systems-level exploration of autonomous driving technology, framed through the lens of cyber-physical systems (CPS). Covering the full autonomy stack, sensing, perception, prediction, planning, control, and actuation, the book integrates computational, physical, and embedded control aspects essential for designing robust, real-time autonomous systems. With in-depth discussions on sensor modalities (LiDAR, RADAR, cameras, GNSS, IMUs), deep learning-based perception algorithms (YOLO, DETR, Mask R-CNN), multi-modal sensor fusion, and control strategies (PID, MPC, Stanley, Pure Pursuit), it bridges academic theory and industrial deployment. Practical insights into simulation tools like CARLA, drive-by-wire systems, and real-world deployment challenges make this book an essential resource for graduate students, researchers, and engineers seeking to understand and build intelligent, adaptive autonomous vehicle systems.
- Comprehensive coverage of the full autonomous vehicle stack: sensing, perception, prediction, planning, control, and actuation.
- In-depth analysis of sensor modalities, including LiDAR, RADAR, cameras, GNSS, and IMUs, with real-world benchmarks and datasets.
- State-of-the-art deep learning algorithms for perception, including YOLO, DETR, Mask R-CNN, and multi-object tracking methods like DeepSORT and ByteTrack.
- Detailed exploration of multi-modal sensor fusion strategies, addressing synchronization, redundancy, and alignment challenges.
- Comparative study of control algorithms, including PID, MPC, Stanley, and Pure Pursuit, with practical tuning strategies.
- Practical insights into simulation frameworks (e.g., CARLA), sensor calibration, and drive-by-wire systems using CAN protocols.
- Deployment-focused content, including sensor mounting, data logging, and validation workflows.
This book is for researchers and industry professionals in robotics, embedded systems, autonomous driving, and cyber-physical systems.