Handling Sensing Failures in Autonomous Mobile Robots

Abstract

This article details the SFX-EH architecture for handling sensing failures in autonomous mobile robots. The SFX-EH uses novel extensions to the generate-and-test method to classify failures with only a partial causal model of the sensor/environment/task interactions for the robot. The generate-and-test methodology exploits the ability of the robot as a physically situated agent to actively test assumptions about the state of sensors, condition of the environment, and validity of task constraints. The SFX-EH uses the type of failure to determine the appropriate recovery strategy: reconfiguration of the logical sensor or logical behavior, recalibration of the sensor or actuator, and corrective actions. The system bypasses classification if all hypotheses lead to the same recovery strategy. Results of the SFX-EH running on two robots with different sensor suites and tasks are presented, demonstrating intelligent failure recovery within a modular, portable implementation.

Global methodology in control architecture to improve mobile robot reliability

“…In [5], the authors use a particle filter based state identification method to detect hardware faults such as battery voltage drops or motor encoder decoupling. Other approaches exist for example, in [6], a model based diagnosis approach is described using a probabilistic hybrid automaton to model the considered failure modes and the nominal mode, or in [7], who propose a "generate and test approach" to check all possible sensing failure origins of current symptoms. However, multiple model oriented approaches may be hampered by state space handling problems when the number of treated faults increases, especially in an embedded and real-time context.…”

“…For example, concerning hardware faults in [6], the authors determined whether a robot should reconfigure, use a de- graded mode or stop on the basis of qualitative constraints on robot components and diagnostic results. In [7], they use exception handling to recover from a detected failure.…”

Model-based fault diagnosis and reconfiguration of robot drives

“…But so far it was only applied for diagnosis of faults in the robots hardware and no repair actions are derived. Rule-based approaches were proposed by Murphy and Hershberger [14] to detect failures in sensing and to recover from them. Additional sensor information were used to generate and test hypotheses to explain symptoms resulting from sensing failures.…”

Fault tolerance enhancement using autonomy adaptation for autonomous mobile robots

“…Fault detection can be done using timing checks, reasonableness checks, safety-bag checks, or model-based monitoring and diagnosis [14]. Some architecture focus on hardware faults, as the SFX-EH [15] which proposes to recover from sensing faults using hardware reconfiguration. Brandstötter et al expose in [16] a model-based fault diagnosis and reconfiguration framework using a probabilistic hybrid automaton.…”