Recovery and reliability prediction in fault tolerant automotive embedded system

The objective of the work is to propose a recovery and reliability prediction (R&R) in automotive embedded system. The existing reliability enhancement models are emphasizing various redundancy techniques both in hardware and software of recovery time minimization from the affected or degraded states in the automotive systems. The recovery oriented computing is based not only on the severity of the damage caused by the uncertain stimuli or random events on system components, important steps towards defining standard-based middleware which can satisfy real-time requirements in an embedded system. These requirements can only be fulfilled if the middleware utilizes the features of a real-time network. The Controller Area Network (CAN) is one of the most important networks in the field of real-time embedded systems. Automotive vehicles through embedded systems for reasons like emission control, vehicle connectivity, safety and cooperative behaviors. As the development often involves stakeholders from different engineering disciplines and organizations, the complexity due to shared requirements, interdependencies of data, functions, and resources, as well as tight constraints in regards to timing safety. A Redundancy Enabled Collaborative Recovery model is proposed in which the correct and timely activation of multiple recovery mechanisms through time and data redundancy techniques are used for the enhancement of system reliability.