The profound impacts of driverless vehicles technology could change to our society remarkably, not to mention the significant enhancements they could bring to the overall safety, efficiency, and convenience of transportation and transit systems. This paper employs deep neural networks to predict the steering angle and throttle values for an autonomous vehicle by obtained images taken from different viewpoints. The proposed convolutional neural network is able to extract the features from the images and find the dependencies for forecasting the steering angle and the speed to keep the vehicle running at the center of the lane automatically. The synthetic images used in our work is generated from Udacity platform.
|Title of host publication||2020 Australian and New Zealand Control Conference (ANZCC)|
|Publication status||Published - Nov 2020|