Session: 14-04-01: Reliability and Safety in Transportation Systems

Paper Number: 69308

Start Time: Tuesday, 10:55 AM

69308 - Design of an Efficient, Low-Cost, Stationary LiDAR System for Roadway Condition Monitoring 

Light Imaging Detection and Ranging (LIDAR) systems were first utilized in aerospace applications in the 1960’s and later employed by NASA as an improvement on radar systems that could not develop high resolution images. LIDAR systems generate point cloud imagery by using laser beams to measure distance to a surface and then combine thousands of points to create a three-dimensional (3-D) image. Since early adaptations of the technology, LIDAR is now common in aerial and sub-terranean geographical surveying and autonomous vehicle operations. The transportation industry uses LIDAR to monitor roadway quality, which can allow hazardous roadway corrosion to be spotted and repaired before endangering drivers. The leading issue with LIDAR availability is the respectively steep price point for effective systems, therefore preventing widespread usage. Developing an inexpensive, efficient, portable, and simple LIDAR device will benefit society. Previous work at the authors’ university in this area involved an initial design that generated high-resolution 3-D imagery but was faulted by limited portability and a long run-time. A second design built upon the initial work by creating a more mobile system with greater processing power but failed to create accurate data and was difficult to use due to issues with gimbal translation and bugs in the C++ programming code.

This effort revisits these previous efforts and implements new hardware and software to increase usability while offering multiple levels of resolution. This design combines a new touchscreen Graphical User Interface (GUI) with an existing rangefinder (Garmin LIDAR-Lite v3HP) that is powered by Raspberry Pi 4 Model B hardware. The rangefinder is housed in a 3-D printed gimbal mount that translates via two stepper motors (28BYJ-48 Small Reduction Stepper Motor) and driver board (Qunqi L298N). The system runs via a new Python script that allows the user to select varying levels of resolution on the GUI prior to data collection. Like the previous two designs, data output is in Cartesian coordinates through a .xyz file format with a MATLAB script used to create a point cloud and two-dimensional image with a depth gradient. This design allows files to be saved onto a Secure Digital flash drive for simple transportation of data. This system can be used as a stationary device that can create 3-D imagery to monitor roadway corrosion or non-commercial use while accessible at a reasonable price point. Overall, this design process created a more efficient, easier to use, and higher accuracy LIDAR system that offers various resolution levels for under the cost of $500.

Presenting Author: Christopher Depcik University of Kansas

Authors:

Jarod Bennett University of Kansas
Mather Saladin University of Kansas
Daniel Sizoo University of Kansas
Spencer Stewart University of Kansas
Graham Wood University of Kansas
Thomas DeAgostino University of Kansas
Christopher Depcik University of Kansas