Development of object detection algorithm for automotive using 3D solid-state lidar sensor

Date
Feb. 2021 - Aug. 2021

Project Objective

β€’
Develop an object recognition model suitable for a fixed 3D lidar sensor being developed for mobility.
β€’
Develop an object recognition model that operates in real-time in an embedded environment
β€’
Establish a pathway to sales to potential buyers of future 3D solid-state lidar and perception application software

Contributions

β€’
Collected, cleaned, processed, and trained 3D lidar datasets from driving environments
β€’
Developed lidar object recognition module in an embedded environment (Raspberry Pi 4)
β€’
Evaluated lidar object recognition algorithm performance
β€’
Evaluated lidar object recognition algorithm speed

Demo Video

β€’
Rule-based Lidar Object Recognition Module
β€’
Learning-based lidar object recognition module