It is necessary to track system state during robotic manipulation. System state is defined to be manipulator and environment configuration. Without tracking, the robot is ignorant of the outcomes of its actions. State tracking enables the robot to respond to sub-task failures. Tracking system state is difficult for complex robots that incorporate hundreds of individual sensor signals. It is difficult to determine which signals are relevant and which are not, and to find low dimensional representations of system state.

The goal of this invention is to provide a tactile feedback loop that can be used to adapt an automated sequence. This invention provides a framework for tracking the state of a robotic system and operating environment during manipulation. This additional layer of logic can enable the robot to execute an arbitrary number of execution paths based on the outcome of its current actions. This knowledge can be used to wrap adaptive control around a task or sequence.

This system enables the robot to monitor system state during task execution. It is easy to teach system states during the offline learning phase. This technology is an enabler for correcting subtask failures.

This work was done by Robert Platt and Joseph Pfeiffer of Johnson Space Center, Adam Sanders and Nathaniel Quillin of General Motors Corporation, and Frank Permenter of Oceaneering. For further information, contact the JSC Technology Transfer Office at (281) 483-3809. MSC-25149-1