For supporting NASA's Robotics, Tele-Robotics and Autonomous Systems Roadmap, we are proposing the "Evolutionary Autonomous Health Monitoring System" (EAHMS) for planetary exploration, which will provide an integral flexible diagnostics and prognostics framework by advanced and novel methods for determining the operational condition in on-board sensors (odometry), actuators, and power systems. In EAHMS, high performance diagnostic techniques provide a foundation for tailoring robust and accurate failure detection and identification (FDI) in key components of a robotic vehicle's locomotion system (e.g. motors, encoders, etc.). This foundation is comprised of innovative and advanced features including: (a) an enhanced collaborative learning engine (eCLE); (b) sensor health diagnostics with slippage awareness based on an Extended Kalman filter sensor fusion process; and (c) an integral system design for optimized reliability. In particular, the eCLE provides a mechanism for facilitating autonomous operation, since it includes self-learning capability. The eCLE is developed within the context of health monitoring, but will also have the capability to be applied to different domains. Another innovation is the support for electronic circuits and boards considering radiation effects.