RU2722258C1 - Autonomous unmanned underwater vehicles - Google Patents

Abstract

FIELD: underwater vessels.SUBSTANCE: invention relates to creation of autonomous unmanned underwater vehicles (AUUV), more specifically their control systems. Autonomous unmanned underwater vehicle (AUUV) comprises an integrated control system (ICS), including combined technical means and subsystems, namely, propelling-and-steering, diving-surfacing, navigation, hydro-acoustic illumination of environment, hydro-acoustic communication and radio communication. All said hardware and subsystems are equipped with their own control means and are interconnected via a data exchange network. Integrated control system includes a scheduling subsystem connected to the data communication network.EFFECT: higher efficiency of controlling AUUV while reducing cost of developing AUUV control algorithms and software implementing them.3 cl, 3 dwg

Description

The invention relates to the field of creating autonomous uninhabited underwater vehicles (AUV), and more precisely, their control systems.

AUV is a fully automatic technical system designed to perform various tasks under water. Since AUV operates in a complex underwater environment, it is equipped with a wide range of technical equipment. The basic set of these tools includes:

- a source of electrical energy (usually a rechargeable battery);

- power steering system;

- diving-surfacing system;

- navigation sensors: inertial navigation system (ANN), echo sounder, log;

- hydroacoustic means of lighting the environment (emitting and receiving antennas, generators of emitted signals, signal processing equipment from the output of the receiving antennas);

- sonar communication equipment;

- radio communication facilities;

- managing computer.

Depending on the purpose of the AUV, the above list can be expanded with additional means of providing movement, sensors for measuring parameters of various physical fields, various manipulators.

All of the listed technical means, as a rule, were created by different developers, have different computing means and different power supply systems. Their information pairing with the host computer, sources and consumers of information is carried out, as a rule, using different data exchange protocols.

In [2], the operation of the control unit of an unmanned aerial vehicle (UAV) for the selection of a flight route is described. The UAV flight route is determined dynamically during the flight or in advance of the flight using navigation data. Based on these data, several route options are constructed that are ranked according to the selected criteria.

In [3], an operating system is described that is focused on applications in robotic systems. This operating system provides the development, configuration, assembly, construction, deployment and management of robotics.

The control of known AUVs (like UAVs), as a rule, is multi-objective [5-7]. This means that the control of the functioning of all the technical means (subsystems) of the AUV is assigned to the control computer, in whose memory a formalized mission plan is laid. In the process of completing the mission, the managing computer in accordance with the mission plan issues control commands to the appropriate technical means of the AUV. The work of the AUV hardware is limited to the execution of commands received from the host computer.

An example of multi-object control is the AUV control system (SU): Autonomous mission controller described in [4] and schematically shown in FIG. 1. Management in this SU is carried out at three levels. 1st level - organization (organization and management), 2nd level - analysis of received data and coordination (supervision and coordination), 3rd level - executive (execution and control). The first and second levels, at which management plans are developed, are implemented in SU ANPA. The third level is implemented in the AUV subsystems. The mission plan and the databases needed to manage it are uploaded onshore to SU. The mission plan is divided into stages, which, in turn, consist of segments. The description of each segment includes: its purpose, a list of AUV subsystems involved in the implementation of this segment, and other data. When fulfilling the current segment of the mission plan, the SUAU issues control commands to the corresponding subsystems of the ANAU and receives messages from them about the results of the teams. Based on these messages, the control unit decides on the possibility of moving to the next segment of the plan. It should be especially noted that all AUV subsystems are implemented on their own hardware and software and are not structurally included in the control system (in Fig. 1, the control system is a rectangle limited by a dashed line). The communication between the AUV subsystems and the control system is carried out using the data exchange bus. The role of AUV subsystems in the management process is limited to the execution of commands received from the control system.

The practice of using the AUV shows that for the successful completion of the mission, the AU AU must not only issue commands in accordance with the formalized plan of the mission, but also fend off all unforeseen contingencies due to both external and internal reasons. External causes of emergency situations include: deliberate and unintentional counteraction from other objects, fixed and mobile obstacles, unforeseen changes in flow, water temperature, sonar conditions, etc. Internal causes are: hardware malfunctions, software failures, inability of individual devices to fulfill the issued him a command due to limited functionality, or the exhaustion of the charge of the power source. All these reasons, as a rule, lead to the deviation of the AUV from the exact execution of a formalized mission assignment. This is manifested in deviation from the route, delay, or the inability to complete individual stages of the mission. As a result, the SUAAU must make very non-trivial decisions that ensure the implementation of the mission plan in unpredictable conditions.

The practice of creating and using AUVs has shown that it is almost impossible to develop such a complex centralized management system. In view of this, most of the AUVs created in the world have very limited functionality.

As a prototype we choose AUV described in [1]. The control circuit of the AUV prototype is shown in FIG. 2. SU ANPA, implemented at the program level, in accordance with the mission plan stored in its memory, at the times indicated in the plan, issues the control commands indicated in the plan to the indicated technical means. The technical tool that received the command executes it and the execution results are reported by the host computer.

In order for the multi-objective SU of ANPA to be effective, when issuing commands to the ANPA subsystems, it must take into account their technical condition, energy and time resources, as well as be able to optimize their functioning in the current situation. To create such a control system, as practice has shown, is not possible due to its high complexity.

Solved technical problem - increasing the effectiveness of the implementation of the AUV missions.

EFFECT: increased management efficiency of AUV while reducing the cost of developing control algorithms and software that implements them.

The specified technical result is achieved by moving from the centralized (multi-object) control of the AUV to decentralized (multi-agent). This means that an integrated management system (IMS) is being created, in which all the above-mentioned technical equipment acquiring the status of agents are included in the form of subsystems, i.e. independent subsystems with their own controls. The ANA SU, which took place in the multi-object version, is replaced by the planning subsystem in the multi-agent version, the functions of which include organizing the implementation of the mission plan by issuing the ISU subsystems of work plans for the upcoming mission phase that correspond to the ANAU formalized mission plan, as well as updating these plans in case of occurrence unforeseen situations hindering the exact implementation of the AUA mission plan.

The ISU subsystems, having received the work plan at the upcoming stage of the mission, optimize its implementation taking into account the current situation and the existing technical capabilities. If necessary, other subsystems are involved to implement the plan. The specific content of this attraction, taking into account the existing technical capabilities, is determined by the exchange of data between these subsystems. To ensure the exchange of data between the subsystems of the ISU, it is equipped with a data exchange network that allows for the exchange of data using unified network protocols.

Note that each of the subsystems, if necessary, can, in turn, consist of several agents with their own management tools, which allows decomposition of the control algorithm of the subsystem and thereby simplify it.

The structure of the ISU AUV is shown in FIG. 3. ISU ANPA includes subsystems:

- planning;

- navigation;

- lighting;

- sonar communication;

- radio communications;

- engine-steering;

- diving ascent.

At the same time, to reduce the size and energy consumption of the ISU AUV:

- the planning subsystem and controls of all subsystems are implemented at the software level of a single computing device, preferably with a distributed architecture (to increase reliability and simplify software development);

- the power supply of all the technical means of the ISU is carried out through a single power supply system connected to the power source of the AUV.

The functioning of the ISU AUV is organized as follows. On the shore, a mission plan is written in memory with a plan for each stage in memory of a single computing device of the ISU. During the movement of the AUV, the planning subsystem issues plans for the upcoming stage to the controls of the ISU subsystems. The controls of the ISU subsystem, having received a plan for the upcoming stage, develops options for its implementation taking into account available resources, from which it selects the optimal plan that meets some quality criterion (for example, minimum execution time, minimum energy consumption, etc.), and organizes it performance. If necessary, the subsystem management tool engages other ISS subsystems to complete the task, while optimizing the joint work of the subsystems, for which it exchanges the necessary data with them over the network. The subsystem that received the stage plan from the planning subsystem informs the planning subsystem of the results of its implementation.

If the results of the deadlines and completeness fit into the plan for the current stage of the mission, the planning subsystem proceeds to issue plans for the next stage in accordance with the plan of the mission. Otherwise, the planning subsystem adjusts the timing and / or content of the next stage. And so on until the completion of the mission.

Consider the management of the AUV while ensuring its movement along a given route. The specified route of the AUV movement is stored in the memory of the planning subsystem. To control the movement of the AUV along a given route, the planning subsystem receives information from the navigation subsystem about the current estimated coordinates of the AUV, as well as the current estimated circular error of the numbering of coordinates. Based on these data, the planning subsystem estimates the deviation of the AUV position from a given route and, if the deviation exceeds the permissible value, makes a decision on the observance. The planning subsystem (PP) issues a request for observation to the navigation subsystem, the control tool of which decides which method to conduct the observation. When choosing an observational method, the navigation subsystem management tool asks for the technical condition of the radio communication subsystem responsible for the observation using signals from the global satellite navigation system (GPS) and the radio navigation system (RNS) signals, as well as the hydroacoustic communication subsystem responsible for observing the signal from the bottom beacons -responders. Also, the navigation subsystem asks the engine-steering subsystem for the availability of resources for the ascent of the AUV to conduct observations on the signals of GSNS and RNS. As a result, the choice of the observation method is carried out taking into account the answers received, as well as data on the equipment of the navigation area using RNS and bottom transponder beacons. If the observational method associated with the floatation of the AUV is selected, the navigation subsystem reports this planning subsystem, which, taking into account the presence of the electric power reserve, issues an application for the ascent of the propulsion-steering subsystem. When the ascent is completed, the planning subsystem reports this to the navigation subsystem, which issues an application to the radio communication subsystem for the observation. The navigation subsystem reports the obtained observation results to the planning subsystem and uses it to correct the numbering.

Thus, the main difference between the multi-agent construction of the ISA APS from the multi-object is the presence of controls in each subsystem of the AUV, which makes their behavior intelligent.

The advantages of multi-agent construction of ISU AUV in comparison with multi-object are:

- increasing the management efficiency of AUVs by optimizing the management algorithms of each subsystem and AUV as a whole;

- reducing the complexity (and, consequently, the price) of creating control algorithms for AUVs and software that implements them;

- reducing the complexity of working out the interaction of all subsystems of the AUV;

- the possibility of creating ISU, unified for various types of AUV.

The latter advantage is especially important, given that the development of control algorithms for technical means is the most time-consuming part of creating an AUV for any purpose.

To confirm the effectiveness of the claimed invention, the operation of the AUV with the above-described typical set of technical means was carried out while fulfilling the mission of the AUV transition to a point with specified coordinates. At the same time, during the transition, emergency situations were randomly modeled due to both external causes (fixed and moving obstacles, limited possibilities for correcting the current coordinates of the AUV), and internal factors (equipment malfunctions). Modeling was carried out both for the multi-object version of the AUV control and for the multi-agent one.

As a result of 1000 modeling cycles, it was found that the probability of completing a mission in the case of multi-agent control option AUV was 0.93, while in the case of multi-object control option it was 0.76. As the analysis showed, the results are due to the fact that with multi-object management, SUAAU is not in all cases able to correctly assess the available resources and capabilities for fulfilling the current stage of the mission plan and, as a result, optimize the implementation of plans by individual subsystems of the AUV.

In addition, the labor-intensiveness of developing control algorithms for AUVs and software that implements them for both control options was evaluated, which showed that the development of algorithms and software in the case of multi-agent control is on average 40% less labor-intensive than in the case of multi-object control. This is explained by the decomposition of algorithms and software in the case of multi-agent control and the associated possibility of attracting specialists of various profiles and competencies to the development of decomposed algorithms.

It was also shown that by combining all the technical means of the AUV into an ISU, the physical volume of equipment in the AUV case is reduced by 12% (due to the creation of a unified computer system and a unified power supply system).

Thus, the claimed technical result - increasing the management efficiency of AUV while reducing the cost of developing control algorithms AUV and software that implements them - can be considered achieved.

Sources of information:

1. Vaulin Yu.V., Inzartsev A.V., Lvov O.Yu., Matvienko Yu.V., Pavin A.M. Reconfigurable control and navigation system for multifunctional underwater robots // Underwater Research and Robotics, 2017, No. 1 (23), p. 4-13.

2. Patent 10163357 (B2) (USA). Navigation assistance data and route planning for drones.

3. Patent 9195233 (B2) (USA). General purpose robotics operating system

4. Patent 6122572 (A) (USA). Autonomous command and control unit for mobile platform.

5. Rzhevsky G.A., Skobelev P.O. How to manage complex systems? Multi-agent technologies for creating intelligent enterprise management systems // Samara: Etching, 2015, 290 pp.

6. Gorodetsky V.I., Grushinsky M.S., Khabalov A.V. Multi-agent systems (review) // News of Artificial Intelligence, 1998, No. 2, p. 64-116.

7. Marino A., Antonelli G., Aguiar A. Pedro., Pascoal A., Chiaverini S. A decentralized strategy for multi-robot sampling / patrolling: theory and experiments // IEEE Transactions on Control Systems Technology. Vol. 23. No. 1. 2015. P.313-322.

Claims (3)

1. Autonomous uninhabited underwater vehicle (AUV), containing an integrated control system (ISU), including the combined technical means and subsystems: propulsion-steering, diving-surfacing, navigation, sonar lighting, sonar, radio communication, characterized in that all the mentioned technical means and subsystems are equipped with their own control means and are informationally connected to each other via a data exchange network, and the planning subsystem connected to the data exchange network is included in the integrated control system. 2. AUV according to claim 1, characterized in that the planning subsystem and controls of all subsystems of the IMS are implemented at the software level of a single computing device, preferably with a distributed architecture. 3. AUV according to claim 1, characterized in that the power supply of all the technical means of the IMS is carried out through a single power supply system connected to an AUV power source.

Images