Situational planning and operational adjustment of the route of the Autonomous robotic underwater vehicle

Abstract

Currently, missions (tasks) for the underwater robot formed using imperative programming methods (both text and graphic), describing in detail the sequence of robot actions that need performed to achieve the desired goal. At the same time, only the operator of the underwater robot, which makes up the mission, for example, the delivery of cargo to the target point, has an idea of the goal itself. Such technology is effective if the robot's mission carried out within a priori scenario. In other cases, it can either not be executed at all, or it can be executed with large violations and a threat to the safety of the device.
When assessing the effectiveness of an underwater robot, the degree of its information autonomy, i.e. the ability to act independently in an unknown or insufficiently defined environment, is of fundamental importance. Therefore, the "intellectualization" of the Autonomous control system of the underwater robot is extremely important for the mission in unforeseen circumstances. For this propose to use intelligent decision support system. Two ways to implement optimal decision-making strategies based on the mathematical apparatus of the theory of Markov and semi-Markov processes using the Bellman optimality principle propose. The considered ways of implementation of optimal strategies of decision - making process relate to the strategy for a short finite time of cargo delivery, which is the most common in practice, and for a long interval of cargo delivery relative to the entire task. In addition, the article discusses ways to find optimal strategies when the time of making single decisions is fixed or when the time of translation is implement randomly.
Hence, the situational approach to decision-making in the planning of the route ARPA is very relevant and allows not only to assess the possible situation on the route, but also to determine the control solutions for the operational adjustment of the route using the intelligent decision support system (ISPR). The development of models of the routing process based on the representation of the situational model in the form of nodes of the graph, the transitions of which correspond to the control solutions.
The paper proposes two ways to implement optimal strategies of decision - making based on the mathematical apparatus of the theory of Markov and semi-Markov processes using the Bellman principle of optimality.