Honors College Theses

Publication Date

2024

Major

Mechanical Engineering (B.S.)

Document Type and Release Option

Thesis (open access)

Faculty Mentor

Valentin Soloiu

Abstract

Distracted driving poses a significant safety hazard and will only exacerbate as the number of modern-day distractions increases. To mitigate this problem, Advanced Driver Assistance Systems (ADAS) features, such as blind spot detection, have been pivotal for the safer operation of vehicles. Towards the same objective, the goal of this research is to utilize 2D LiDAR sensors to create a blind spot detection system that will detect objects and surfaces that are outside of the driver’s field of view. A comparative analysis was conducted by developing a 2D LiDAR-based system utilizing NVIDIA Jetson Orion Nano and Python alongside an ultrasonic-based system using Arduino Mega 2560 and an HC-SR04 sensor. Field trials were conducted at speeds of ten and fifteen miles per hour, revealing that the current 2D LiDAR system falls short compared to the ultrasonic counterpart. Specifically, the average method displayed inaccuracies, while the lowest distance method failed to return to the initial state after the target vehicle passed. Future enhancements are proposed, including code optimization and debugging. Transitioning to C++ could potentially increase the speed of the 2D LiDAR system, while improved debugging may enhance system reliability. These optimizations promise to render the 2D LiDAR system viable for advanced driver assistance systems (ADAS).

Thesis Summary

The study explored the potential of utilizing 2D LiDAR technology for enhancing blind spot detection systems compared to traditional sensors. A 2D LiDAR-based system was developed and tested alongside an ultrasonic-based system. Trials conducted at varying speeds indicated that the current state of the 2D LiDAR system falls short in accuracy and reliability compared to the ultrasonic system. Suggestions for future optimization include code refinement and debugging, potentially transitioning to C++ for improved real-time performance and ensuring the reliability of detection methods.

Download

Share

COinS