Consumer interest for autonomous vehicles is growing around the world. Formal verification techniques are needed for thorough verification and validation of such safety-critical systems. Applying static verification techniques for such complex systems that are also increasingly designed and developed using Artificial Intelligence based approaches is challenging and has limitations. In this work, we propose the use of light-weight dynamic formal verification approaches, runtime verification and

more »

... nforcement. We prototype a self-driving car and propose to apply runtime monitoring to ensure safety of the vehicle. For the development of the prototype, we use Raspberry pi as a master device and Arduino Uno as a slave for steering the vehicle. We use various image processing methods to develop a working hardware prototype model. The output of the controller is fed to the monitor (generated using formal runtime monitor synthesis approach), which enforces desired safety policies on the output of the system. We propose that these formal dynamic monitoring approaches can also be used on Neural Network based controllers. The developed hardware model can act as a test bed to illustrate practical applicability of formal runtime monitor synthesis theory and tools in the context of cyber-physical systems such as autonomous vehicle.