Hyun-Wook Jo

Metric based nonblocking supervisory control of discrete event systems

This paper considers metric based nonblocking supervisory control of discrete event systems (DESs). We introduce the definition of metric indicating distace from one state to other state. The states are paritioned into nonblocked states and blocked states based on metric. Then, we suggest the control law for supervisor which blocks blocks the event sequences from nonblocked states to blocked states. Moreover, we prove that the supervisory controller following the control law is nonblocking when the conditions are satisfied. Through the simple memory example, we show that the trade-off between efficiency and robustness can be easily controlled.

Task planning for service robots with optimal supervisory control

The problem of task planning for multiple service robot using supervisory control of discrete event systems is considered. We apply the optimal supervisory control technique to obtain the cost efficient task planning without degrading the quality of service. Specifically the cost function is assigned to each state transition. Then, we introduce an algorithm to minimize the maximum cost. Through an example, we verify the usefulness of our algorithm.

Completeness of Scalable Path Planning for Multiple Robots using a Graph of Shared Cycles

Many path planning methods have been studied to find collision-free paths for multiple robots in complex environment. However, the previous methods cannot guarantee completeness or scalability in the finding the paths. In this paper, we propose the method transforming a roadmap to the graph of cycles having more than one shared edge with other cycles. Then, we show that all multiple robots can be moved to their goal positions without collision using the graph. The proposed algorithm guarantees not only the completeness, but also the scalability. Especially, we show that the proposed algorithm provides efficient paths of all robots than existing methods and can be available for the roadmap having extremely many robots.

Optimal supervisory control under partial observation for service robots

In this paper, we propose an optimal supervisory control algorithm for discrete event system (DES) to minimize the cost function under partial observation. Under the partial observation, it is generally impossible to know the current state. Moreover, it can cause the cost overestimation which prevents the optimal supervisory control. Specifically, we introduce the notions of multiple cost events and state indices. We propose a new cost-calculating algorithm based on the notions to resolve the overestimation by identifying the current state. To validate the proposed algorithm, we present an example of the task planning of the practical service robot.