Development of a vehicle robotic driver with intelligent control system modelling for automated standard driving-cycle tests
New road vehicles are required to undergo several specific tests to meet the requirement set by governing bodies in various markets. These tests are often carried out over specific driving-cycles. To carry out lab-based driving-cycle tests, a typical vehicle manufacturer will employ a trained driver...
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | English English English |
| Published: |
2019
|
| Subjects: | |
| Online Access: | http://eprints.uthm.edu.my/460/1/24p%20AMMAR%20ABDO%20MOHAMMED%20MAHDI.pdf http://eprints.uthm.edu.my/460/2/AMMAR%20%20ABDO%20MOHAMMED%20MAHDI%20COPYRIGHT%20DECLARATION.pdf http://eprints.uthm.edu.my/460/3/AMMAR%20%20ABDO%20MOHAMMED%20MAHDI%20WATERMARK.pdf |
| Summary: | New road vehicles are required to undergo several specific tests to meet the requirement set by governing bodies in various markets. These tests are often carried out over specific driving-cycles. To carry out lab-based driving-cycle tests, a typical vehicle manufacturer will employ a trained driver to follow driving profiles on a chassis dynamometer. This project involves development of a robotic driver controller for the automation of dynamometer-based vehicle testing according to industry standard driving cycle tests and produce repeatable results by replacing the traditional method of employing a human driver with a robot driver. The throttle and brake pedals control systems modelling and design for automatic transmission vehicle are implemented, with Fuzzy model reference adaptive control (Fuzzy MRAC) as the main controller. The vehicle model was developed using black-box modelling approach where simulations are performed based on real-time data and processed using Matlab System Identification tool. The Fuzzy MRAC was then designed within the simulations to attain the driving performance. The vehicle model response was sent as feedback to the robotic DC linear actuator motor which was modelled based on DC linear actuator motor design specification. The results obtained from simulation and modelling experiment were discussed and compared. The performed work concludes that system identification modelling with best fit accuracy of 79.93% can be applied in Fuzzy MRAC to ensure smooth and accurate vehicle driving pattern behavior even when the leading vehicle exhibits highly dynamic speed behavior during driving-cycle test. The performance of the vehicle model has shown an average 0.07 MSE for the throttle system and 0.008 MSE for the brake system of the vehicle model. |
|---|