Level-3 Autonomous Driving System by using GTA V: Open Source
Mohan Sai Tanishq A1, G. Suresh2, B. Uday Kiran Reddy3
1Mohan Sai Tanishq A, Department of Electronics and Communication Engineering, KL Deemed to be University, Vijayawada (Andhra Pradesh), India.
2G.Suresh, Department of Electronics and Communication Engineering, KL Deemed to be University, Vijayawada (Andhra Pradesh), India.
3B.Uday Kiran Reddy, Department of Electronics and Communication Engineering, KL Deemed to be University, Vijayawada (Andhra Pradesh), India.
Manuscript received on 07 April 2019 | Revised Manuscript received on 20 April 2019 | Manuscript published on 30 April 2019 | PP: 1257-1260 | Volume-8 Issue-6, April 2019 | Retrieval Number: F3872048619/19©BEIESP
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: To delineate bare pixels from a single front camera to steering command, by training a convolutional neural network (CNN). Basically a CNN is a type of end-to-end approach. With moderate amount of training data, the neural network was simulated in Virtual environment, and it even performed well on the roads except on roads with no lane marking. With the help of Fuzzy logic and Nvidia neural network our model performed exceptionally well even in heavy traffic conditions. Also we used Fuzzy Logic for speed control. We used GTA V as the simulation environment. The recommended hardware required is 6GB GTX 1060, and a 250GB high speed SSD with 3.5 GB/s read speed, 1.5 GB/s write speed. The system operates at 10 frames-per-second (FPS).
Keyword: Artificial Intelligence, Deep learning, Machine Learning, Neural Network.
Scope of the Article: Systems and Software Engineering