Decision-Making for Remote Control of Emergency Mobile Robot

The proposed paper analyses the decision-making for remote control of emergency mobile robotic platforms. There is described a model of robot’s decision-making, based on predicate logics, which combines local and remote procedures of goals and actions determination. The developed model uses the description of local emergency workspace added with remote-supplied information, based on distributed computer vision system. In core of decision-making system is problem-solving engine, which includes the set of action procedures, goals, specifications of robot and its sensor system. The procedure of action typically contains the result of action, the set of subgoals to solve, set of added facts and removed facts on workspace. Facts of considered workspace includes the descriptions of objects inside workspace with their coordinates, dimensions, and probable identification. To describe the emergency workspace here is proposed information model, based on set theory descriptions. While emergency robot is mostly considered as autonomous system with independent decision-making, the role of remote control is still essential for fast recognition of unknown or damaged objects, fast generation of movements and actions in emergency workspace, fast solution on actions in dangerous situations, for fast information support from other robots and people. The key role of remote control system can be executed by human operator with help of intelligent tools of other remote control systems, which provide monitoring of emergency workspace, based on computer-vision of acting robot and other information sources. The proposed ideas can be considered as a basis for real emergency robot control system implementation.