Cyber-Physical Systems (CPS) are subject to various faults due to failing actuators, sensors or structural components. The increasing size and complexity of modern systems result in cost-and time-intensive manual fault handling. To enable systems to adapt to faults autonomously, reconfig-uration, i.e. the identification of a new valid configuration that recovers operation, is necessary. This paper presents an extension of the recently published reconfiguration algorithm AutoConf and the application to the Environmental Control and Life Support System (ECLSS) of the COLUMBUS module aboard the ISS. The implementation draws on a qualitative system model formulated in propositional logic. The corresponding satisfiability problem is solved by a state of the art SAT-Solver. The extension consists of three contributions, namely a health status implementation , a dynamic causal graph and a problem-specific formulation of serial dependencies of the actuators. Both a static and dynamic evaluation (integrated simulation) of the extended reconfig-uration algorithm is presented for 72 fault cases, covering a wide range of faults.    «

Cyber-Physical Systems (CPS) are subject to various faults due to failing actuators, sensors or structural components. The increasing size and complexity of modern systems result in cost-and time-intensive manual fault handling. To enable systems to adapt to faults autonomously, reconfig-uration, i.e. the identification of a new valid configuration that recovers operation, is necessary. This paper presents an extension of the recently published reconfiguration algorithm AutoConf and the applicat...    »