A method and device for autonomous decision-making of cross-domain maneuvering mode of amphibious vehicle

Abstract

The present invention discloses a method and device for autonomous decision-making of cross-domain working condition maneuvering mode of amphibious vehicles, the method comprising the following steps: S1. Receiving relevant data from various sensor devices and control modules of the platform through interfaces and preprocessing; S2. Transmitting the data obtained in step S1 to the cross-domain decision controller for analysis, and generating the optimal candidate working condition corresponding to the current driving environment; S3. Converting the optimal candidate working condition generated in step S2 into a control instruction for switching the maneuvering mode of the maneuvering actuator, and outputting it to the vehicle control system to complete the maneuvering mode switching process of the amphibious vehicle. The method and device of the present invention support the autonomous switching capability of the cross-domain maneuvering mode of the amphibious vehicle autonomously by virtue of its high accuracy and high adaptability.

Description

Technical Field

[0001] The present invention relates to the technical field of vehicle engineering and shipbuilding industry, and in particular to a method and device for autonomously making decisions on cross-domain working condition maneuvering modes of amphibious vehicles. Background Art

[0002] An amphibious vehicle is a special vehicle that has both water and land driving functions. Its engine power is distributed and switched through a transfer case, so that it can have power on water, power on land, and both power working at the same time. Special vehicles that can complete functions such as water cruising and offshore landing play an irreplaceable role in unique geographical environments.

[0003] However, due to the different working requirements of amphibious vehicles in water and on land, the driving mode of amphibious vehicles is determined by the current environmental conditions. When performing cross-domain tasks, amphibious vehicles in the prior art mainly rely on the decision of switching maneuvering modes performed by the driver or remote control personnel based on a variety of perception factors. The specific judgment is based on coastline, water depth, wading depth, geographical map, vehicle posture, typical targets in water or land, etc., or the vehicle realizes the perception and decision-making capabilities of the above data by configuring various sensors, but it only realizes the fusion and judgment of various types of data, and often cannot make accurate cross-domain autonomous switching control, and cannot achieve true autonomy.

[0004] Therefore, in view of the fact that existing amphibious vehicles are unable to achieve accurate and autonomous switching between amphibious modes, it is necessary to invent a method for an amphibious vehicle to autonomously switch between maneuvering modes under cross-domain conditions, which can autonomously and accurately change its own driving mode according to the surrounding environment, so as to further improve the autonomy and adaptability of amphibious vehicles to the environment. Summary of the invention

[0005] In view of the problems existing in the prior art, the present invention aims to complete the switching of mobile working conditions by maintaining the traditional collection of sensor data, determining the current working conditions by switching criteria and outputting them to the vehicle electrical system, and then studying the autonomous decision-making and control of Petri nets, building a cross-domain environmental characteristic parameter model and an autonomous decision-making controller based on Petri nets, and providing the Petri net with judgment elements of transition through the decision maker, so that the Petri net can autonomously determine the working condition judgment conditions and switching time, assist the traditional cross-domain changing working condition process to make autonomous decisions, and provide an autonomous decision-making method for the cross-domain working condition maneuvering mode of amphibious vehicles. With its high accuracy and high adaptability, it supports the autonomous switching capability of amphibious vehicles to autonomously plan cross-domain maneuvering modes.

[0006] In order to achieve the above object, the present invention provides an autonomous decision-making method for a cross-domain working condition maneuvering mode of an amphibious vehicle, the method comprising the following steps:

[0007] S1. Receive relevant data from various sensor devices and control systems on the platform through a data interface and pre-process it to generate a set of environmental characteristic parameters; the various sensor devices include at least a liquid level sensor, a depth sensor and a visual sensor;

[0008] S2. The parameter set obtained in step S1 is passed to the cross-domain decision controller for analysis to generate the optimal candidate operating condition of the environment in which the vehicle is traveling, and the influence of the closed-loop control is applied to the next planning cycle;

[0009] S3. Convert the optimal candidate operating condition generated in step S2 into a control instruction for switching the maneuvering mode of the maneuvering actuator, and output it to the vehicle control system to complete the maneuvering mode switching process of the amphibious vehicle.

[0010] Further, in step S1, the data interface includes: a vehicle status information interface, a navigation information interface and an environmental information interface; the data information includes: vehicle status information, navigation information and environmental information; the preprocessing method includes: missing value processing and outlier processing.

[0011] Further, in step S2, the cross-domain decision controller is deployed in the cross-domain autonomous decision module. On the basis of the cross-domain environmental characteristic parameter model, the autonomous decision-making idea based on Petri net is added. According to the input environmental data and the changes of the environmental medium under the cross-domain, the cross-domain working condition is divided into different segments. Based on the different environmental characteristics, structural characteristics and kinematic characteristics of amphibious vehicles in different working condition segments, relevant data is collected and characteristic parameters are selected therefrom to optimize and upgrade the current working condition judgment, and output the optimal working condition that the vehicle should be in under the current environment.

[0012] Further, in step S3, the control instructions include: start and stop instructions for the land thruster and start and stop instructions for the water thruster.

[0013] Furthermore, in step S1, the specific execution flow of input data preprocessing is as follows:

[0014] During real-time operation, the vehicle collects and analyzes various data through the data interface, including: vehicle status information collected by the vehicle control system, the vehicle status information includes vehicle speed, acceleration, and / or motor actuator status; position and attitude information collected by the navigation module, the position and attitude information includes vehicle positioning and attitude and odometer; liquid level environment information collected by the liquid level sensor; water depth environment information collected by the depth sensor; environmental information collected by the visual sensor, the environmental information includes grid map information and target recognition information; the collected data is parsed and preprocessed to create a set of environmental feature parameters for the processed data, thereby completing the collection and analysis of the environmental data.

[0015] Furthermore, in step S2, the specific execution process of cross-domain working condition decision is as follows:

[0016] The parameter set obtained from S1 is used to input the cross-domain decision controller to perform cross-domain environmental analysis. After completion, the optimal candidate operating condition that the vehicle may be in under this parameter set information is output, which is one of the land operating condition, the water operating condition, and the water and land operating condition, where the water and land operating condition includes shallows or swamps.

[0017] Furthermore, in step S3, the specific execution process of the vehicle maneuvering switch is as follows:

[0018] Based on the optimal candidate operating condition obtained from S2, the optimal candidate start-stop state corresponding to the maneuvering actuator is determined based on this operating condition and output to the vehicle control system as a maneuvering switching instruction. The vehicle control system executes the start and stop of the corresponding mechanism according to the output instruction, thereby completing the maneuvering mode switching process of the amphibious vehicle.

[0019] On the other hand, the present invention provides an autonomous decision-making device for the cross-domain working condition maneuvering mode of an amphibious vehicle, and the device is used to implement the autonomous decision-making method for the cross-domain working condition maneuvering mode of an amphibious vehicle according to the present invention.

[0020] Furthermore, the device includes a vehicle control system, a vehicle maneuvering actuator, a cross-domain working condition decision module, a navigation module, a liquid level sensor, a depth sensor and a visual sensor.

[0021] Furthermore, the vehicle control system is used to control the core system of vehicle operation and can obtain real-time vehicle operation status data; the maneuvering actuator is used for vehicle power drive; the cross-domain working condition decision module is used to determine whether the vehicle performs cross-domain maneuvering switching; the navigation system is used to obtain vehicle position, posture and odometer information; the liquid level sensor is used to obtain liquid level information; the depth sensor is used to obtain water depth information; the visual sensor is used to obtain grid map information and target recognition information.

[0022] The beneficial effects of the present invention are as follows:

[0023] (1) The autonomous decision-making method for the cross-domain maneuvering mode of an amphibious vehicle of the present invention solves the problems that manual operation is required during the cross-domain maneuvering switching process, the switching point is difficult to determine during the autonomous switching, and the switching criteria are imperfect. The maneuvering mode switching in the cross-domain environment is completed through a series of sensing devices and intelligent algorithms, which solves the dependence of the vehicle on the operator during the land-water switching, improves the autonomy of the amphibious vehicle, and further improves the adaptability of the amphibious vehicle to work in complex environments.

[0024] (2) The autonomous decision-making method for the cross-domain maneuvering mode of an amphibious vehicle of the present invention is not limited to a specific type of vehicle, but can be deployed on a variety of amphibious platforms. It has a high degree of scalability and can meet the diverse needs of amphibious operations. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 It is a flow chart of the present invention;

[0026] Figure 2 This is a diagram of the autonomous decision-making process for the vehicle's cross-domain operating condition maneuvering mode according to the present invention. DETAILED DESCRIPTION

[0027] The technical solution of the present invention will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0028] In the description of the present invention, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc., indicating the orientation or positional relationship, are based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first", "second", and "third" are used for descriptive purposes only, and cannot be understood as indicating or implying relative importance.

[0029] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0030] The following combination Figure 1 , Figure 2 The specific embodiments of the present invention are described in detail. It should be understood that the specific embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0031] The present invention provides an autonomous decision-making device for cross-domain working condition maneuvering mode of amphibious vehicles, the device comprising a vehicle control system, a maneuvering actuator, a cross-domain working condition decision-making module, a navigation system, a liquid level sensor, a depth sensor and a visual sensor. Among them:

[0032] Vehicle control system: a core system used to control vehicle operation, which can obtain real-time vehicle operation status data, including vehicle speed, acceleration, and motor actuator status;

[0033] Motor actuator: A device used for vehicle power drive that can change the start and stop of the motor by receiving instructions from the vehicle control module;

[0034] Cross-domain working condition decision module: used to send instructions to the vehicle control system on whether to execute cross-domain maneuver switching;

[0035] Navigation system: used to obtain vehicle positioning information, posture information and odometer information;

[0036] Liquid level sensor: used to obtain liquid level information;

[0037] Depth sensor: used to obtain water depth information;

[0038] Visual sensor: used to obtain raster map information and target recognition information;

[0039] like Figure 1 As shown, the autonomous decision-making method for the cross-domain maneuvering mode of an amphibious vehicle provided by the present invention mainly comprises the following steps:

[0040] S1. Receive relevant data from various sensor devices and control systems on the platform through data interfaces and perform preprocessing to generate a set of environmental characteristic parameters. Specifically, vehicle status information (P1), navigation information (P2), and environmental sensor information (P3) are obtained by the vehicle control system, navigation module, and environmental module; the above data are preprocessed (t1) to obtain a set of environmental characteristic parameters (P4);

[0041] S2. The parameter set obtained in step S1 is passed to the cross-domain decision controller for analysis, and the optimal candidate operating condition of the vehicle's driving environment is generated, and the influence of closed-loop control is applied to the next planning cycle; specifically, the environmental characteristic parameter set (P4) is passed to the cross-domain environmental decision module for analysis (t2), and the optimal candidate operating condition (P5) is determined; the optimal candidate operating condition is added to the cross-domain environmental decision module for closed-loop detection (t3) for auxiliary analysis.

[0042] S3. Convert the optimal candidate operating condition generated in step S2 into a control instruction for switching the motorized mode of the motorized actuator, and output it to the vehicle control system to complete the motorized mode switching process of the amphibious vehicle. Specifically, convert the optimal candidate operating condition into a mode switching instruction of the motorized actuator (t4); send the instruction to the vehicle control module, including the switching instruction of the crawler track (P6) and the switching instruction of the hydraulic propeller (P7).

[0043] In step S1, during real-time operation, the vehicle collects and analyzes various data through the data interface, including: vehicle status information such as vehicle speed, acceleration, and motor actuator status collected by the vehicle control system; vehicle location, attitude, and odometer and other position and attitude information collected by the navigation module; liquid level environment information collected by the liquid level sensor, water depth environment information collected by the depth sensor, grid map information and target recognition information collected by the visual sensor and other environmental information. The collected data is analyzed and preprocessed to obtain more standardized and clean data to improve data quality and accuracy, and a set of environmental feature parameters is created for the processed data to complete the collection and analysis of environmental data.

[0044] In step S2, the parameter set obtained from S1 is input into the cross-domain decision controller for cross-domain environmental analysis. After completion, the optimal candidate operating condition that the vehicle may be in under this parameter set information is output, which may be one of land operating conditions, water operating conditions, and water and land operating conditions (such as shallows, swamps, etc.).

[0045] In step S3, based on the optimal candidate operating condition obtained from S2, the optimal candidate start-stop state corresponding to the motorized actuator is determined based on this operating condition and output to the vehicle control system as a motorized switching instruction. The vehicle control system will execute the start-stop of the corresponding mechanism according to the output instruction to complete the motorized mode switching process of the amphibious vehicle.

[0046] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention is fully and clearly described below by taking an amphibious diesel tracked vehicle as an example:

[0047] For data interaction between modules, this method uses Json string for reception and transmission. Json string is a way to transmit data information in the form of Key-Value. Although different programming languages ​​have different basic data types, strings are common to all programming languages. Based on this feature, Json string adopts the form of "{"parameter name":"parameter value"}" to transmit information in different language systems. In the application process, each sub-module serializes the data information into Json string and passes it to this algorithm. After receiving the Json string, this algorithm deserializes it into the required data information and passes it to the main body of the algorithm. Its advantages are that the information transmitted is flexible, the serialization and deserialization process is relatively simple and fast, which reduces the difficulty of development and can also reduce unnecessary system overhead during data parsing. And because all the information of Json string is in the form of key-value pairs, it only needs to add or modify the information of key-value pairs when maintaining or expanding data information, which is convenient for expansion and maintenance. Although Json string adopts the method of plain text transmission, as far as the current system is concerned, all data transmission is carried out within the system, and the information will not be exposed to the outside, and there is no security concern.

[0048] For data preprocessing, since the acquired raw data often cannot be used directly for analysis, there will be various problems, such as invalid information, duplicate values, missing values, outliers, etc., which need to be cleaned and converted to become more suitable for subsequent work. This method mainly performs noise elimination operations, by analyzing the collected information and performing noise detection. Specifically, the attribute "noise weight" is added to all dimensional data, and a weighted method is used to determine the noise data source in a certain state, and the weight is dynamically adjusted from 0 to 10 according to the current state. The higher the weight, the greater the interference of the current attribute on the data analysis, and it can be excluded during the data analysis process, filtering out data that will reduce the system operation efficiency and increase resource overhead, and eliminating irrelevant data.

[0049] During the operation of the vehicle, the vehicle body changes with the environmental medium, and the vehicle posture, force, power and other data will change accordingly. The data output by the various sensors of the vehicle will change under different media and environmental conditions. From the perspective of mobility mode, there are three main mobility modes for amphibious vehicles: track drive mode when moving on land, water jet propulsion device drive mode when moving on the water surface, and both drive modes work simultaneously when entering the water and landing on offshore shallows.

[0050] Regarding the construction of a cross-domain decision controller, since the cross-domain work of amphibious vehicles is one of their normal working scenarios, there is a data judgment and autonomous decision-making process in the cross-domain work process of the tank from water surface to land and water surface, which includes continuously changing controlled quantities and discrete events such as switches and selections. Therefore, the variable working condition system is a typical process-based autonomous judgment control system and a typical hybrid system. Petri net is an important modeling and analysis tool for describing hybrid systems, which can describe and analyze system characteristics such as concurrency, conflict, synchronization and resource contention. In view of the problem that the switching point is difficult to determine and the switching criteria are imperfect during the switching process of variable working conditions, an autonomous decision controller based on Petri net is built for the environmental characteristic parameter model, and the model is added to the Petri net system. The controller provides the judgment element of the transition for the Petri net. In each cycle, according to the values ​​of the current moment and the previous moment, the conditions and switching timing for executing the maneuvering switch are determined to complete the decision of the cross-domain working condition maneuvering mode autonomous decision.

[0051] According to the definition of the cross-domain working conditions, working medium and environment of amphibious vehicles, the working conditions of technical decisions are determined, which are mainly state sets and event sets established based on vehicle working status information, navigation information and environmental information. The library model places and transfer environments are matched, and the library places and transitions corresponding to the Petri net are designed according to the vehicle status information, navigation information, environmental information, etc. of the amphibious vehicle to establish the Petri net model diagram. The execution process description is shown in Figure 2, the execution process is as follows:

[0052] The specific implementation process is as follows:

[0053] (a) Vehicle status information (P1), navigation information (P2), and environment-related sensor information (P3) are obtained by the vehicle control system, navigation module, and environment module;

[0054] (b) performing the preprocessing of the above data (t1) to obtain a set of environmental characteristic parameters (P4);

[0055] (c) The environmental characteristic parameter set (P4) is passed to the cross-domain environmental decision module for analysis (t2) to determine the optimal candidate working condition (P5);

[0056] (d) On the one hand, the optimal candidate working condition is added to the cross-domain environmental decision module for closed-loop detection (t3) for auxiliary analysis;

[0057] (e) on the other hand, converting the optimal candidate operating condition into a mode switching command of the motorized actuator (t4);

[0058] (f) Sending instructions to the vehicle control module, including a track switching instruction (P6) and a hydraulic thruster switching instruction (P7).

[0059] Among them, a and b correspond to step S1, c and d correspond to step S2, and e and f correspond to step S3.

[0060] The information input by the input interface of the present invention includes: vehicle status information, position and posture information, and environmental sensor information; the information output by the output interface includes a device power mode switching instruction. Figure 2 The symbols in the table mean:

[0061]

[0062] In a specific embodiment, taking visual sensor information as an example, the execution process of generating the optimal candidate operating condition of the environment in which the vehicle is traveling is as follows:

[0063] Embodiment: Determining the optimal candidate operating condition of an amphibious vehicle according to the operating condition environment information of the vehicle:

[0064] (a) Visual information acquisition and information matrix generation

[0065] For visual sensors, they can collect a variety of information around the vehicle. The grid map is used to describe the point cloud data of the area measured by the laser radar, including the position, distance, occupancy and other information relative to the vehicle. The target recognition information is used to describe the image data of the area measured by the camera, including the shape and material of the vehicle. The above information is sorted, classified, identified, screened and fused to obtain a collection of visual environment information;

[0066] For any point in the visual environment information set collected by the visual sensor, the grid map information of the point is used to obtain the distance of the target object relative to the vehicle. ,angle and occupy ; Using the target identification information of this point, you can get its detailed information and get the distance of the target object ,angle , Material ;

[0067] Will , , , , , By summarizing and fusion the data, a collection of visual feature information can be obtained. ,in is the occupancy information of the vehicle’s surroundings, and S is the material information of the vehicle’s surroundings;

[0068] For depth sensors, the water depth around the vehicle can be collected ;

[0069] For liquid level sensors, the liquid level around the vehicle can be collected ;

[0070] Summarize the above data to get the environmental characteristic parameter matrix ;

[0071] (b) Feature extraction and cross-domain rule identification

[0072] For each of the above environmental characteristic parameters, there is a critical value corresponding to the environmental change. By fusing all characteristic parameters and comparing them with the critical value at the cross-domain moment, calculating and fusing the results, the actual value of the matching degree between the working environment and the land environment can be obtained. , this value is affected by all characteristic parameters and is the cumulative value of the matching degree between all characteristic parameters and the terrestrial environment;

[0073] When the above characteristic parameters are more in line with the visual characteristics of the target in the land environment, when the target material is road, water depth , Liquid level In this case, the characteristic parameters match the terrestrial environment to a high degree. At a high level;

[0074] By verifying the changes in various characteristic parameters during the alternation between water and land, a critical value is set for the matching degree between the working environment and the land environment ,when , it is considered that the possibility of the operating environment being land is higher than expected, and the optimal candidate operating condition is output as land operating condition.

[0075] (c) Dynamic decision-making of working conditions

[0076] For the optimal candidate operating condition obtained through the cross-domain controller, the corresponding execution state instruction is output to the vehicle control system. If the optimal candidate operating condition is the land operating condition, the vehicle control system is output with the instruction to start the land thruster and stop the water thruster; if the optimal candidate operating condition is the water operating condition, the vehicle control system is output with the instruction to stop the land thruster and start the water thruster; if the optimal candidate operating condition is the water-land operating condition, the vehicle control system is output with the instruction to start the land thruster and start the water thruster. In this way, the maneuvering mode switching process of the amphibious vehicle is completed.

[0077] Any process or method description in the flowchart of the present invention or described in other ways herein can be understood as a module, segment or part of a code including one or more executable instructions for implementing the steps of a specific logical function or process, which can be implemented in any computer-readable medium for use by an instruction execution system, device or equipment, and the computer-readable medium can be any medium containing storage, communication, propagation or transmission programs for use by execution systems, devices or equipment, including read-only memories, magnetic disks or optical disks, etc.

[0078] In the description of this specification, the description with reference to the terms "embodiment", "example", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. In addition, those skilled in the art can combine or combine different embodiments or examples described in this specification and the features therein without causing any contradiction.

[0079] Although the above content has shown and described the embodiments of the present invention, it can be understood that the above embodiments are exemplary and cannot be understood as limitations of the present invention. Ordinary technicians in the field can change, modify, replace, modify and other update operations on the above embodiments within the scope of the present invention.

Claims

1. An autonomous decision-making method for cross-domain maneuvering mode of an amphibious vehicle, characterized in that: The method comprises the following steps: S1. Receive relevant data from various sensor devices and control systems on the platform through a data interface and pre-process it to generate a set of environmental characteristic parameters; the various sensor devices include at least a liquid level sensor, a depth sensor and a visual sensor; S2. The parameter set obtained in step S1 is passed to the cross-domain decision controller for analysis to generate the optimal candidate operating conditions of the environment in which the vehicle is traveling, and the influence of the closed-loop control is applied to the next planning cycle; the optimal candidate operating conditions include three types: land, water and water; S3. Convert the optimal candidate operating condition generated in step S2 into a control instruction for switching the motor mode of the motor actuator, and output it to the vehicle control system to complete the motor mode switching process of the amphibious vehicle; Among them, the optimal candidate working condition of the amphibious vehicle is determined according to the set of environmental characteristic parameters of the vehicle: (a) Visual information acquisition and information matrix generation The visual sensor is used to collect information around the vehicle. The grid map is used to describe the point cloud data of the area measured by the laser radar, including the distance, angle, and occupancy relative to the vehicle. The target recognition information is used to describe the image data of the area measured by the camera, including the distance, angle, and material relative to the vehicle. The collected information is sorted to obtain a set of visual environment information. For any point in the visual environment information set collected by the visual sensor, the grid map information of the point is used to obtain the distance of the target object relative to the vehicle. ,angle and occupy ; Use the target recognition information of this point to get the distance of the target ,angle , Material ; Will , , , , , Aggregate and fuse the data to obtain a set of visual feature information ,in is the occupancy information of the vehicle’s surroundings, and S is the material information of the vehicle’s surroundings; For the depth sensor, the water depth around the vehicle is collected ; For the liquid level sensor, the liquid level around the vehicle is collected ; Summarize the data to get the environmental characteristic parameter matrix ; (b) Feature extraction and cross-domain rule identification For each environmental characteristic parameter, there is a critical value corresponding to the environmental change. By fusing all environmental characteristic parameters and comparing them with their cross-domain critical values, calculating and fusing the results, the actual value of the matching degree between the working environment and the land environment is obtained. , actual value Affected by all environmental characteristic parameters, it is the cumulative value of the matching degree between all environmental characteristic parameters and the terrestrial environment; When the environmental characteristic parameters are more in line with the visual characteristics of the target in the land environment, when the target material is road, water depth , Liquid level In this case, the environmental characteristic parameters match the terrestrial environment to a high degree. At a high level; By verifying the changes in various environmental characteristic parameters during the alternation between water and land, a critical value is set for the matching degree between the working environment and the terrestrial environment ,when , it is considered that the possibility of the operating environment being land is higher than expected, and the optimal candidate operating condition is output as land operating condition.

2. The autonomous decision-making method for cross-domain maneuvering mode of an amphibious vehicle according to claim 1 is characterized in that: In step S1, the data interface includes: a vehicle status information interface, a navigation information interface and an environment information interface; the data information includes: vehicle status information, navigation information and environment information; the preprocessing method includes: missing value processing and outlier processing.

3. The autonomous decision-making method for cross-domain maneuvering mode of an amphibious vehicle according to claim 1 is characterized in that: In step S2, the cross-domain decision controller is deployed in the cross-domain autonomous decision module. The autonomous decision-making idea based on Petri net is added on the basis of the cross-domain environmental characteristic parameter model. According to the input environmental characteristic parameter set and the environmental changes under the cross-domain, the working condition is divided into different segments. Based on the different environmental characteristics, structural characteristics and kinematic characteristics of amphibious vehicles in different working condition segments, relevant data is collected and environmental characteristic parameters are selected therefrom to optimize and upgrade the current working condition judgment, and output the optimal working condition that the vehicle should be in under the current environment.

4. The autonomous decision-making method for cross-domain maneuvering mode of an amphibious vehicle according to claim 1 is characterized in that: In step S3, the control instructions include: start and stop instructions for the land thruster, and start and stop instructions for the water thruster.

5. The autonomous decision-making method for cross-domain maneuvering mode of an amphibious vehicle according to claim 2 is characterized in that: In step S1, the specific execution flow of input data preprocessing is as follows: During real-time operation, the vehicle collects and analyzes various data through the data interface, including: vehicle status information collected by the vehicle control system, the vehicle status information includes vehicle speed, acceleration, and / or motor actuator status; position and attitude information collected by the navigation module, the position and attitude information includes vehicle positioning and attitude and odometer; liquid level environment information collected by the liquid level sensor; water depth environment information collected by the depth sensor; environmental information collected by the visual sensor, the environmental information includes grid map information and target recognition information; the collected data is parsed and preprocessed to create a set of environmental feature parameters for the processed data, thereby completing the collection and analysis of the environmental data.

6. The autonomous decision-making method for cross-domain maneuvering mode of an amphibious vehicle according to claim 3 is characterized in that: In step S2, the specific execution process of cross-domain working condition decision is as follows: The parameter set obtained from S1 is used to input the cross-domain decision controller to perform cross-domain environmental analysis. After completion, the optimal candidate operating condition that the vehicle may be in under this parameter set information is output, which is one of the land operating condition, the water operating condition, and the water and land operating condition, where the water and land operating condition includes shallows or swamps.

7. The autonomous decision-making method for cross-domain maneuvering mode of an amphibious vehicle according to claim 4 is characterized in that: In step S3, the specific execution process of the vehicle maneuvering switch is as follows: Based on the optimal candidate operating condition obtained from S2, the optimal candidate start-stop state corresponding to the motorized actuator is determined based on this operating condition and output to the vehicle control system as a motorized switching instruction. The vehicle control system executes the start-stop of the motorized actuator according to the output instruction to complete the motorized mode switching process of the amphibious vehicle.

8. An autonomous decision-making device for cross-domain maneuvering mode of an amphibious vehicle, characterized in that: The device is used to implement the autonomous decision-making method for the cross-domain maneuvering mode of an amphibious vehicle according to any one of claims 3 or 6.

9. The autonomous decision-making device for cross-domain maneuvering mode of an amphibious vehicle according to claim 8 is characterized in that: The device includes a vehicle control system, a vehicle maneuvering actuator, a cross-domain autonomous decision-making module, a navigation module, a liquid level sensor, a depth sensor and a visual sensor.

10. The autonomous decision-making device for cross-domain maneuvering mode of an amphibious vehicle according to claim 8, characterized in that: The vehicle control system is used to control the core system of vehicle operation and to obtain real-time vehicle operation status data; the motorized actuator is used for vehicle power drive; the cross-domain autonomous decision-making module is used to determine whether the vehicle performs cross-domain motorized switching; the navigation module is used to obtain vehicle position, posture and mileage information; the liquid level sensor is used to obtain liquid level information; the depth sensor is used to obtain water depth information; and the visual sensor is used to obtain grid map information and target recognition information.

Images