Demonstrating planning work usually depends on hardware demos that reviewers cannot actually try. Educational comparisons between A*, Dijkstra, and sampling-based planners are rare in interactive form.
Loading portfolio
Robotics and computer vision / 2024
Robot Path Planning Simulator
Demonstrating planning work usually depends on hardware demos that reviewers cannot actually try. Educational comparisons between A*, Dijkstra, and sampling-based planners are rare in interactive form.
30×20 grid canvas with obstacle / erase / start / goal modes, algorithm picker, run / clear / maze / reset controls, and a stats panel covering path cost, visited cells, runtime, and success. Pure-Python planner modules on the backend with deterministic unit tests; mirrored TypeScript implementations for instant client-side interaction.
- Role
- Designer and engineer (solo)
- For
- Independent or academic work
- Stage
- Maintained
My part
What I built
- A*, Dijkstra, and RRT-style planners with shared module boundary
- Visited-cell visualisation makes algorithmic differences readable
- API validation and search-expansion limits prevent runaway requests
Current stage and limits
- Grid-based. No continuous collision checking or footprint inflation.
- No kinematic constraints, dynamic obstacles, or localisation uncertainty.
- Educational artefact; not a substitute for a full robotics navigation stack.
Technical detailOpen the system view
SYSTEM ARCHITECTURE
Path Planning Simulator system view
30×20 grid canvas with obstacle / erase / start / goal modes, algorithm picker, run / clear / maze / reset controls, and a stats panel covering path cost, visited cells, runtime, and success. Pure-Python planner modules on the backend with deterministic unit tests; mirrored TypeScript implementations for instant client-side interaction.30×20 grid canvas with obstacle / erase / start / goal modes, algorithm picker, run / clear / maze / reset controls, and a stats panel covering path cost, visited cells, runtime, and success. Pure-Python planner modules on the backend with deterministic unit tests; mirrored TypeScript implementations for instant client-side interaction.
A*, Dijkstra, and RRT-style planners with shared module boundary
Visited-cell visualisation makes algorithmic differences readable
API validation and search-expansion limits prevent runaway requests
- 01Input / actorconnects toInterface
- 02Interfaceconnects toCore service
- 03Core serviceconnects toOperational boundary
- 04Operational boundaryconnects toState / outcome
Built with
Links
More work
ARIRA ROS-based autonomous retail robot combining real-time shelf perception, multi-object tracking, autonomous navigation, customer-request handling, and ERP synchronisation.Validated in simulationRobot Guide DogReal-time ROS perception and audio-guidance system developed for a Unitree Go1 guide-dog concept, combining object detection, scene context, and synchronised spoken cues.Research prototypeVision Operations PlatformBrowser-to-inference platform for streaming camera frames to a FastAPI detection service, visualising detections in the browser, and monitoring streaming behaviour in real time.Maintained