Investigation of Applying Verification Methods to a Cyber-Physical System

When developing a cyber-physical system, designers mainly concentrate on functionality and dynamics.  As algorithms become more complex, multiple computers are needed to run the full system safely. This may result in undesirable outcomes such as an increase in cost or processing power. In this research, verification methods were used to produce safe code for the CAT Vehicle, the autonomous vehicle being developed at the University of Arizona. The verification methods ensure that the network of the autonomous vehicle runs within four constraints. The four constraints are cost, processing power, bandwidth, and latency. Operating within these constraints allows the car to maximize its data processing potential. A Domain Specific Modeling Language (DSML) was designed using a Web-based Generic Modeling Environment to produce a model of the network. The DSML uses interpreters to verify the dynamic constraints and generate template code. If constraints are violated, technique selectors suggest the user to rewire connections or exchange a computer within the network. When all constraints are met, the information can be sent quickly and accurately to allow the vehicle to operate in real time. The purpose of this paper is to present a Domain Specific Modeling Language with interpreters containing verification methods to allow the production of safe and efficient code that can run the network system of the CAT Vehicle.