A real-time vision-guided robotics project integrating computer vision, embedded control, and custom mechanical hardware to enable human-following behavior.

This project explores how camera-based perception can be integrated with embedded motor control to drive powered motion on an existing chair platform. The system detects and tracks a human target in real time and commands motor outputs accordingly, emphasizing safe operation and system stability.

• Perception: Python + OpenCV and MediaPipe for real-time human detection and position estimation
• Control: Proportional control loop converting target position into motion commands
• Embedded: Arduino-based motor control driving high-current DC motors via IBT-2 motor drivers
• Mechanical: Custom CAD-designed, 3D-printed motor mounts integrated onto an existing chair frame

• Processes camera input to estimate relative target position
• Applies proportional control to compute speed and direction commands
• Implements speed limiting to maintain stable behavior

• Receives serial commands from the vision pipeline
• Drives DC motors through IBT-2 drivers
• Enforces motion smoothing and safety limits

• Designed and 3D-printed custom motor mounts
• Integrated motors without modifying the original chair structure

• Maintaining stable tracking under variable lighting conditions
• Synchronizing real-time vision processing with embedded actuation
• Managing torque and power constraints to avoid stalling or overheating
• Ensuring safe behavior during continuous operation

The project demonstrates an end-to-end robotics system combining perception, control, and hardware integration, emphasizing practical engineering tradeoffs and system reliability.