Enhancing the mobility performance of Off-road Autonomous Unmanned Ground Vehicles (UGV) by increasing their response to their surrounding environment

Abstract

Unmanned Ground Vehicles (UGVs) provide an ideal, long-range approach to a variety of basic issues for military and defence and/or civil security forces: performing in a hostile area or investigating potentially hazardous items or circumstances without injuring soldiers. A UGV has many of the essential capabilities of the person it replaces to perform those missions. Specifically, it needs to be able to navigate on the basis of stored maps and real-time visual data, interpret its environment as a matter of course (obstacle and danger detection and avoidance). The aim of this these is to enhance the mobility performance for Off-road autonomies ground vehicle. Hence, two important aspects of the autonomous vehicles were discussed, firstly vehicle prediction which is the most important aspect to the autonomous vehicles. A several comparisons of sensors and different machine learning algorithms were analysed a line with two case studies evaluation hence, a fused sensor is required to achieve an efficient vehicle prediction. Secondly, from the mobility perspective, the aim of the any UGVs mission is succeed in driving form A to B, not only because they need to be more responsive, however, also, the must be highly mobile. Therefore, the vehicle platform and the vehicle classification were discussed. Hence, as the entire mission of the vehicle would be in soft surface a tracked vehicle platform would be used to achieve a better mobility performance and easily overcoming the obstacles compared with wheeled vehicle.