Vehicle model for the Wisconsin Autonomous Chevrolet Bolt
Problem statement
The current Wisconsin Autonomous Chevrolet Bolt requires a validated method to predict and deliver the torque necessary to maintain a desired speed under varying steering inputs. The program is unable to smoothly control the car during its course, leading to jerky and unstable movements. Despite placing 2nd overall, this problem was observed during the SAE Autodrive Challenge II, the vehicle’s inconsistent low-speed control reduced performance can affect Wisconsin Autonomous’ credibility, lowering the morale of both the Wisconsin Autonomous Team along with its advisor, who put so much effort into the program. Aside from a competition environment, this also further validates the fears of the general public regarding the safety and reliability of Autonomous Vehicles as a whole, both of which show that this is a pressing problem urgently requiring a solution. The car will be driven and rigorously tested in multiple environments, primarily in environments with speeds under 15 kph and gradients+/- 6 percent, which simulate a variety of environments that an Autonomous car may be in, in order to ensure that the car will perform reliably. While some people may attribute the weak performance of the Wisconsin Autonomous Vehicle to a technical/software issue, the club advisor had mentioned that numerous software models created by students in the past were ineffective and while they may have worked in a theoretical environment under ideal conditions, they did not fare well in real world environments, showing the need for a new mechanical perspective. As a team of mechanical engineers with experience in vehicle dynamics, data logging, and modeling, we are very committed to solving this problem. We intend to use the data collected in design formulas and tables, which were collected from Human Driving Data Logs to provide the Autonomous team with the information necessary in order for them to program a car which can drive smoothly and which is successfully able to tackle any obstacle in its path.
Team members
Liam Jones – accountant
William Pfeifle – communicator
Praket Akshantala – admin
Drew Rosh-Gorsky – facilitator
Client
Glenn Bower
UW – Mechanical Engineering