Identifying Model-Based Reconfiguration Goals through Functional Deficiencies

Model-based diagnosis is now advanced to the point au- tonomous systems face some uncertain and faulty situations with success by correctly identifying sets of potential component faults. The next step toward more autonomy is to have the system recover- ing itself after faults occur, a process known as model-based recon- figuration. After faults occur, given a prediction of what the nominal expected state of the system should be and the belief state that results from the diagnosis operation, this paper shows how to automatically determine the functional deficiencies induced by the faults. These de- ficiencies are characterized in the case of uncertain state estimates. A methodology is then presented to determine the reconfiguration goals from the deficiencies. Finally, a recovery process interleaves planning and model predictive control to restore the lost functionali- ties in prioritized order.