Autonomous Navigation of an Electric All-Terrain Vehicle Along Waypoint-Defined Trails

Search for this publication on Google Scholar
Daniel, J. (2022). Autonomous Navigation of an Electric All-Terrain Vehicle Along Waypoint-Defined Trails. Unc Charlotte Electronic Theses And Dissertations.
Download PDF

Analytics

65 views ◎
12 downloads ⇓


Abstract

Previous research attempted to autonomously control an internal combustion all-terrain vehicle platform but was unsuccessful due to the inability to reliably control speed. To address this, an electric motor conversion is performed on the platform to enhance its stability and controllability. A robust trail detection methodology is applied to this electric all-terrain vehicle (eATV) prototype to autonomously traverse a 3- to 5-meter-wide wooded trail, an environment that introduces unique challenges for autonomous vehicle control. The vehicle can follow a predetermined route defined by periodic waypoints placed by a human along a trail. A suite of real-time image processing algorithms is developed to respond to input from an Intel RealSense depth camera mounted on the platform. Various simultaneous solutions for path following are prioritized using a confidence scoring approach. Image processing and control techniques are also introduced for obstacle detection and avoidance schemes. The prototype vehicle has successfully navigated waypoint routes along walking trails for over 500-meters while detecting obstacles and pedestrians in real-time. To accommodate global positioning system denial that occurs occasionally in wooded environments, algorithms based on image processing are implemented for navigation toward the subsequent waypoint. An adaptive throttle-braking algorithm is also introduced to maintain a targeted velocity throughout the traversed environment. This research investigates and fulfills a budget-conscious methodology for autonomous off-road vehicle navigation along trails.

Details
Author

Daniel, Jacob

Title

Autonomous Navigation of an Electric All-Terrain Vehicle Along Waypoint-Defined Trails

Physical Description

1 online resource (153 pages) : PDF

Date

2022

Degree Granting Institution

University of North Carolina at Charlotte

Abstract

Previous research attempted to autonomously control an internal combustion all-terrain vehicle platform but was unsuccessful due to the inability to reliably control speed. To address this, an electric motor conversion is performed on the platform to enhance its stability and controllability. A robust trail detection methodology is applied to this electric all-terrain vehicle (eATV) prototype to autonomously traverse a 3- to 5-meter-wide wooded trail, an environment that introduces unique challenges for autonomous vehicle control. The vehicle can follow a predetermined route defined by periodic waypoints placed by a human along a trail. A suite of real-time image processing algorithms is developed to respond to input from an Intel RealSense depth camera mounted on the platform. Various simultaneous solutions for path following are prioritized using a confidence scoring approach. Image processing and control techniques are also introduced for obstacle detection and avoidance schemes. The prototype vehicle has successfully navigated waypoint routes along walking trails for over 500-meters while detecting obstacles and pedestrians in real-time. To accommodate global positioning system denial that occurs occasionally in wooded environments, algorithms based on image processing are implemented for navigation toward the subsequent waypoint. An adaptive throttle-braking algorithm is also introduced to maintain a targeted velocity throughout the traversed environment. This research investigates and fulfills a budget-conscious methodology for autonomous off-road vehicle navigation along trails.

Genre

masters theses

Subjects--Topics

Engineering
Electrical engineering
Mechanical engineering

Degree

M.S.

Keywords

Autonomy
Depth Imaging
Electric Vehicle
Image Processing
Waypoint Navigation

Subject Area

Applied Energy & Electromechanical System

Advisor(s)

Browne, Aidan

Committee Members

Smith, Michael
Williams, Wesley

Degree Note

Thesis (M.S.)--University of North Carolina at Charlotte, 2022.

Rights Statement

This Item is protected by copyright and/or related rights. You are free to use this Item in any way that is permitted by the copyright and related rights legislation that applies to your use. For other uses you need to obtain permission from the rights-holder(s). For additional information, see http://rightsstatements.org/page/InC/1.0/.

Rights Holder Information

Copyright is held by the author unless otherwise indicated.

Identifier

Daniel_uncc_0694N_13320

Permalink

http://hdl.handle.net/20.500.13093/etd:3152