Remotely Accessible Open Test Platform for CPS Transportation and CAV Research

Abstract

This paper presents the work in progress for the development of the first known remotely accessible open research platform for indoor experimentation in cyber-physical system (CPS) unmanned systems and transportation research. This portion of the work focuses on the establishment of a multi-robot based platform to address preliminary experimentation needs for Connected and Autonomous Vehicle (CAV) research under different scenarios. This testbed, composed of Create platforms at this stage, ultimately aims to address the need for a realistic validation of CAV strategies for maximum traffic throughput and safety under different road, traffic, and environmental conditions as well as different human or autonomous driving behaviors in urban and highway traffic, prior to outdoor experiments. Different from other existing examples of such indoor systems, the individual platforms are equipped with autonomous capabilities that could be used to reflect different driver/autonomous vehicle behaviors, when needed. Also a decentralized control architecture is adopted to better reflect a realistic traffic operation under different static and dynamic conditions of urban and highway traffic, i.e. work zones etc. In that sense, this testbed could also accommodate research for individual autonomous and semi-autonomous vehicles. The developed system is tested under different CAV scenarios, and results are provided that demonstrate the benefits of the Cooperative Adaptive Cruise Control (CACC) strategy.