Physical AI for Robotics and Automation Training in Marina Bay, Singapore

The Physical AI for Robotics and Automation Training focuses on applying Artificial Intelligence to physical systems, enabling robots and automated machines to perceive, decide, and act in real-world environments.

The course covers robot perception, control, learning, simulation, and safety, helping participants design and deploy AI-powered robotic and automation systems for industrial and service environments.

Introduction to Physical AI

• What is Physical AI?
• Differences between digital AI and physical AI
• Role of AI in robotics and automation
• Industrial and service robot use cases

Robotics System Architecture

• Robot hardware and software stack
• ROS and middleware concepts
• Real-time vs non-real-time systems
• Edge vs cloud AI for robots

Sensors & Perception

• Cameras, LiDAR, depth sensors
• Sensor calibration basics
• Perception pipelines
• Object detection and tracking

Computer Vision for Robotics

• Vision models for robots
• Visual localization and mapping (intro)
• Visual inspection use cases
• Vision performance in real environments

Sensor Fusion

• Combining vision, IMU, LiDAR, and encoders
• State estimation basics
• Kalman filter concepts (high-level)
• Robust perception strategies

Motion Planning & Navigation

• Path planning concepts
• Obstacle avoidance
• Localization and mapping (SLAM overview)
• Real-time navigation systems

Control Systems & AI Integration

• Classical control vs learning-based control
• Reinforcement learning for control (intro)
• Hybrid AI + control architectures
• Stability and safety considerations

Manipulation & Grasping

• Robot arms and end effectors
• Grasp planning
• Vision-guided manipulation
• Industrial pick-and-place systems

Simulation & Digital Twins

• Using simulation for training and testing
• Sim-to-real transfer concepts
• Physics engines overview
• Scenario testing and validation

Learning from Demonstration

• Imitation learning basics
• Collecting demonstration data
• Training behavior models
• Human-in-the-loop teaching

Edge AI for Robotics

• Deploying models on edge devices
• Latency and real-time constraints
• Model optimization
• Hardware acceleration (conceptual)

Safety, Reliability & Certification

• Functional safety concepts
• Fail-safe and redundancy
• Testing and validation
• Human-robot interaction safety

Industrial Automation Integration

• PLC and SCADA integration (conceptual)
• Factory systems and MES
• Robot cell integration
• IT/OT convergence

Monitoring & Maintenance

• Health monitoring for robots
• Predictive maintenance concepts
• Logging and diagnostics
• Continuous improvement

AI Ethics & Governance for Physical Systems

• Ethical considerations for physical AI
• Accountability and incident handling
• Regulatory considerations
• Responsible deployment

Hands-on Exercises

Summary and Conclusion