CANTAV: A Cloud Centric Framework for Navigation and Control of Autonomous Road Vehicles

Abstract

Autonomous vehicles are becoming a new trend in automotive industry. The rapid developments in this field portend to a future where current vehicles are replaced by autonomous ones. There is a need for sophisticated and responsive traffic control mechanisms for such future systems. There is also a need for generic framework for developing and studying autonomous road traffic systems. This paper describes the design and development of a Cloud centric Architecture and simulation framework for Navigation and Traffic control of Autonomous Vehicles (CANTAV) for developing navigation and control systems for automated vehicles and vehicular traffic. The framework is based on message passing which is extensible. It is designed to provide facilities for navigating the vehicle, clear traffic congestion by re-routing traffic. As interfacing such an architecture with real world autonomous vehicles is not feasible, and no existing simulation framework supports such an architecture, an agent based simulation framework, CANTAV simulation framework, is developed for simulation and visualisation of road networks and autonomous vehicular traffic on them. The developed system was tested and validated successfully using several scenarios. Performance of the system is analysed from the system and vehicle perspectives for different transportation network and vehicle arrangements in the system.