Unintentional lane departure accidents are one of the biggest reasons for the causalities that occur due to human errors. By incorporating lane-keeping features in vehicles, many accidents can be avoided. The lane-keeping system operates by auto-steering the vehicle in order to keep it within the desired lane, despite of changes in road conditions and other interferences. Accurate steering angle prediction is crucial to keep the vehicle within the road boundaries, which is a challenging task. The main difficulty in this regard is to identify the drivable road area on heterogeneous road types varying in color, texture, illumination conditions, and lane marking types. This strenuous problem can be addressed by two approaches, namely, 'computer-vision-based approach' and 'imitation-learning-based approach'. To the best of our knowledge, at present, there is no such detailed review study covering both the approaches and their related optimization techniques. This comprehensive review attempts to provide a clear picture of both approaches of steering angle prediction in the form of step by step procedures. The taxonomy of steering angle prediction has been presented in the paper for a better comprehension of the problem. We have also discussed open research problems at the end of the paper to help the researchers of this area to discover new research horizons.

No abstract

Deep learning techniques, particularly convolutional neural networks (CNNs), have exhibited remarkable performance in solving visionrelated problems, especially in unpredictable, dynamic, and challenging environments. In autonomous vehicles, imitation-learning-based steering angle prediction is viable due to the visual imagery comprehension of CNNs. In this regard, globally, researchers are currently focusing on the architectural design and optimization of the hyperparameters of CNNs to achieve the best results. Literature has proven the superiority of metaheuristic algorithms over the manual-tuning of CNNs. However, to the best of our knowledge, these techniques are yet to be applied to address the problem of imitationlearning-based steering angle prediction. Thus, in this study, we examine the application of the bat algorithm and particle swarm optimization algorithm for the optimization of the CNN model and its hyperparameters, which are employed to solve the steering angle prediction problem. To validate the performance of each hyperparameters' set and architectural parameters' set, we utilized the Udacity steering angle dataset and obtained the best results at the following hyperparameter set: optimizer, Adagrad; learning rate, 0.0052; and nonlinear activation function, exponential linear unit. As per our findings, we determined that the deep learning models show better results but require more training epochs and time as compared to shallower ones. Results show the superiority of our approach in optimizing CNNs through metaheuristic algorithms as compared with the manual-tuning approach. Infield testing was also performed using the model trained with the optimal architecture, which we developed using our approach.