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.