Modeling and analysis of real-time issues in rehabilitation robotic systems using coloured Petri nets

In this paper, we discuss some of the real-time issues that must be addressed in the design of rehabilitation robotic systems. A Coloured Petri Net (CPN) model is developed to analyze critical timing and resource requirements for achieving reliable and safe performance. A CPU scheduling problem is introduced and analyzed. An example of a robotic system treating injured or impaired limbs is used as a study case. This work shows the advantages of using CPN to model and simulate the robot-control system in real-time. This work is part of a research and development project to design and build an intelligent robot for physical and occupational therapy to help limb recovery. It is assumed that both the position and velocity of the patient´s limb are measurable and controllable in real-time. The therapy process duration and frequency of limb movement are also controlled by the robot-control system. A time concept is introduced to the design to investigate the performance of the control system. Compact and extendable system design is achieved by utilizing the hierarchical aspect of CPN. This paper presents simulation results that demonstrate satisfactory mission critical performance of the system.