AI-Powered Kitchen Assistant Robot
Robotic system integrating ROS (Robot Operating System), computer vision with TensorFlow, and a web interface with VueJS for kitchen task assistance.
The Challenge
Building a cross-platform robotic ecosystem that coordinated hardware control, visual object recognition, and an accessible user interface. The main technical challenge was integrating heterogeneous systems (ROS, web framework, ML models) into a coherent workflow with bidirectional real-time communication.
My Role
As Full Stack and robotics developer, I implemented components across all three ecosystem layers:
- Developed the responsive web interface with VueJS and Bulma for remote robot control
- Integrated TensorFlow models for visual recognition of ingredients and utensils
- Implemented bidirectional real-time communication between the web app and ROS
- Optimized ML models for execution on resource-constrained embedded hardware
Technical Approach
Robotic control:
- ROS for control, coordination, and communication between robot nodes
- ROS node architecture for separation of concerns (navigation, manipulation, perception)
Computer vision:
- TensorFlow for recognition of kitchen ingredients and utensils
- Model optimization for embedded hardware execution
Web interface:
- VueJS + Bulma for responsive remote control web interface
- Python backend for business logic and ROS communication bridge
- Real-time telemetry and robot status monitoring
Challenges Solved
- Integration of heterogeneous systems (ROS, web, ML) into a unified workflow
- Bidirectional real-time communication between web browser and ROS nodes
- TensorFlow model optimization for embedded hardware with memory and CPU constraints
- Synchronization of multiple concurrent subsystems
Outcome
Functional demonstration of robotics, artificial intelligence, and web technologies integration to create a practical home assistant with remote control and visual perception capabilities.