- Edge-to-Cloud Integration: Focuses on low-latency Edge inference and embedded systems hardware critical for vehicle safety.
- Full-Stack Perception: Covers deep learning for object detection, semantic segmentation, and behavioral recognition in dynamic environments.
- Safety-First Framework: Dedicated modules on functional safety (ISO 26262), cybersecurity, and ethical AI accountability.
- Real-World Simulation: Hands-on projects using virtual validation workflows and high-definition (HD) mapping strategies.
- Introduction to AVs and smart transportation ecosystems
- Evolution from ADAS to Level 5 full autonomy
- Operational Design Domains (ODD) and system boundaries
- Role of AI in perception, planning, and control workflows
- Camera, LiDAR, Radar, Ultrasonic, and GPS hardware overview
- Multi-modal sensor fusion for robust environmental understanding
- Sensor calibration, synchronization, and data acquisition
- Perception challenges in adverse weather and dynamic conditions
- Deep learning architectures for mobility systems
- Object detection, lane tracking, and semantic segmentation
- Scene understanding and human behavior recognition
- Reliability, safety, and model evaluation metrics
- Principles of vehicle positioning and localization
- Simultaneous Localization and Mapping (SLAM) techniques
- High-Definition (HD) maps and route planning layers
- Navigation in structured urban and unstructured off-road environments
- Motion planning and trajectory generation algorithms
- Vehicle control: steering, braking, and acceleration feedback loops
- AI decision-making in complex, high-traffic scenarios
- Obstacle avoidance and road-user interaction modeling
- Real-time computing and low-latency architecture
- Embedded AI hardware (NVIDIA Jetson, SoC) and software stacks
- Edge inference optimization and model pruning
- System integration: power, memory, and thermal constraints
- Functional safety and fail-safe system design
- Cybersecurity for connected and autonomous fleets
- Ethical decision-making and AI accountability
- Regulatory frameworks, global standards, and validation
- Case studies: Autonomous cars vs. delivery robots
- AI in driver monitoring and advanced ADAS
- Virtual validation and hardware-in-the-loop (HIL) workflows
- Future: V2X (Vehicle-to-Everything) and connected mobility
Python & C++ for Real-time Systems
Computer Vision: OpenCV, YOLO, Segmentation
CARLA / Gazebo Simulators
TensorRT for Edge Inference
- Automotive engineers and roboticists
- Embedded system developers and AI researchers
- Transportation planners and policy analysts
- Postgraduate students in Data Science, AI, or Automotive Engineering
Prerequisites: Foundational knowledge of Python or C++ and basic linear algebra. Prior exposure to machine learning is beneficial.






