Software Development for High-Speed Autonomous Ground Vehicles
School Name
South Carolina Governor's School for Science & Mathematics
Grade Level
12th Grade
Presentation Topic
Computer Science
Presentation Type
Mentored
Abstract
Self-driving cars are of great interest in computer science and artificial intelligence due to their speed, efficiency, and affordability. Despite the attention, no one has been able to find the “perfect” solution: the implementation of completely predictable and reliable autonomous transportation. This research serves as an introduction to filling that gap by achieving total autonomy on a smaller scale through programming the Massachusetts Institute of Technology’s RACECAR (Rapid Autonomous Complex Environment Competing Ackermann-steering Robot), creating a safe and robust autonomous ground vehicle that will serve as an example for autonomy on passenger-carrying cars. Python, the advanced control systems on the car, ROS (Robot Operating System), and OpenCV were the main tools utilized to enable the car to drive on its own. My team created an autonomous race car capable of following walls, detecting and avoiding obstacles, and changing paths based on visual perception of its environment. These components were combined into a state machine that implemented different controllers and commands based on detection of AR (augmented reality) tags. The car proved able to navigate autonomously through a complex, dynamic course both safely and quickly. Autonomy was achieved, and, in the future, the implemented tactics can be translated to a passenger-carrying car to fulfill the ultimate goal of autonomous research: fully driverless transportation.
Recommended Citation
Smith, Hollis, "Software Development for High-Speed Autonomous Ground Vehicles" (2019). South Carolina Junior Academy of Science. 158.
https://scholarexchange.furman.edu/scjas/2019/all/158
Location
Founders Hall 140 A
Start Date
3-30-2019 11:15 AM
Presentation Format
Oral Only
Group Project
No
Software Development for High-Speed Autonomous Ground Vehicles
Founders Hall 140 A
Self-driving cars are of great interest in computer science and artificial intelligence due to their speed, efficiency, and affordability. Despite the attention, no one has been able to find the “perfect” solution: the implementation of completely predictable and reliable autonomous transportation. This research serves as an introduction to filling that gap by achieving total autonomy on a smaller scale through programming the Massachusetts Institute of Technology’s RACECAR (Rapid Autonomous Complex Environment Competing Ackermann-steering Robot), creating a safe and robust autonomous ground vehicle that will serve as an example for autonomy on passenger-carrying cars. Python, the advanced control systems on the car, ROS (Robot Operating System), and OpenCV were the main tools utilized to enable the car to drive on its own. My team created an autonomous race car capable of following walls, detecting and avoiding obstacles, and changing paths based on visual perception of its environment. These components were combined into a state machine that implemented different controllers and commands based on detection of AR (augmented reality) tags. The car proved able to navigate autonomously through a complex, dynamic course both safely and quickly. Autonomy was achieved, and, in the future, the implemented tactics can be translated to a passenger-carrying car to fulfill the ultimate goal of autonomous research: fully driverless transportation.
