Conceptual Design of an Autonomous Vehicle for Road Anomaly Detection and Manoeuvring

Abstract

The decaying road infrastructure in Nigeria raises the need for autonomous vehicles capable of detecting and navigating road anomalies. Existing control schemes for navigation are unsuitable for implementation due to the unstructured and dynamic nature of the environment. Due to the aforementioned factors, this paper presents a conceptual design of an autonomous vehicle for road anomaly detection and navigation. The vehicle utilizes a computer vision scheme based on Convolutional Neural Networks (CNN) for road anomaly detection and Model Predictive Control (MPC) for manoeuvring. The implementation of a vehicle prototype with remote monitoring and control capabilities is also presented in this paper. The implementation of this will provide an efficient scheme for road anomaly detection and manoeuvring in autonomous vehicles.