- Projects 3
- Followers 1
Saurabh Banga
Pittsburgh, Pennsylvania
ENVIRONMENT MAPPING ROBOT
The project revolves around leveraging a RC car with a fairly decent USB camera. The processing -(inferencing) would be done using Intel's openVINO & object detection using Deep Learning (Convolutional Neural Networks). The latter's processing would be done via the NCS Movidius Stick. ...learn more
Project status: Under Development
RealSense™, Internet of Things, Artificial Intelligence
Intel Technologies
OpenVINO,
Movidius NCS
Overview / Usage
To identify any wild/unclaimed object found lying on the ground. Leverage object detection to identify if it is an average looking suitcase or something suspicious and accordingly take action (raise an alarm with the authorities and tagging it).
As the title suggests, this project would be use for Environment Mapping, which could be leveraged in various situations:
-
Public places like Airports/Stations.
-
Private places like Hotels which regularly require monitoring
Methodology / Approach
This project will show how to transform a (UP2 Board + Camera ) powered RC car into one capable of object detection and autonomous driving. To do this two deep neural networks would be deployed. One for object detection and the other for autonomous driving using inference for steering and throttle. The UP2 serves as the vehicle computer. 2 NCS Sticks would be used for running parallel networks. Pictures attached below represents the intended output I hope to achieve.
Technologies Used
- RC Car
- UP2 Board
- Logitech or RealSense Camera (D435i)
- Intel® OpenVINO
- Intel® NCS 2
Repository
https://github.com/rishabhbanga/Environment-Mapping-Robot
Collaborators