CN114692520B - A multi-scenario unmanned boat virtual simulation test platform and test method - Google Patents

Abstract

The invention discloses a multi-scene-oriented unmanned ship virtual simulation test platform and a multi-scene-oriented unmanned ship virtual simulation test method, wherein the multi-scene-oriented unmanned ship virtual simulation test platform comprises a marine environment simulation module, a test evaluation module, a dynamics simulation module, a sensor simulation module and an autonomous navigation interaction module; the testing method comprises the steps of setting sea state grades and testing items; generating a tested ocean test scene and a related scene, and arranging a tested virtual unmanned ship in the scene; the virtual sensor acquires sensing information and inputs the information into a control terminal of the actual unmanned ship; the control terminal analyzes the sensing data to obtain a control instruction, and transmits the control instruction back to the virtual test platform to realize navigation control of the virtual unmanned ship; after the sailing is finished, the sailing data are recorded, and an evaluation result is generated. The unmanned ship autonomous navigation performance test method can effectively improve the unmanned ship autonomous navigation performance test effect, reduce the test time and reduce the test cost.

Description

A multi-scenario unmanned boat virtual simulation test platform and test method

Technical field

The invention relates to the field of unmanned driving technology, and in particular to a multi-scene oriented unmanned boat virtual simulation test platform and a test method.

Background technique

Surface unmanned vehicles are playing an increasing role in commercial, scientific and military applications and are now widely used in maritime patrol, environmental monitoring, underwater mapping and ocean research. Autonomous navigation systems require extensive training and verification before formal deployment and operation to ensure product stability. Although real-ship testing can truly reflect the performance of unmanned boats, it requires too much manpower and material resources and is generally used as a verification method in the last step of the development process. In order to improve testing efficiency and reduce testing costs, using virtual simulation testing methods before actual testing is a necessary process in development.

However, unlike other unmanned systems, the complexity and changeability of the ocean environment itself greatly increases the difficulty of building a virtual test site for unmanned boats. The existing virtual test fields for unmanned boats mainly have two shortcomings: on the one hand, the existing virtual test fields for unmanned boats generally adopt the testing method of unmanned equipment on land and only simulate motion on a two-dimensional plane. The impact of the marine environment on unmanned vehicles is rarely considered, which is far from the actual complex navigation situation. Starting from the theory of fluid dynamics and using numerical methods to solve the flow field, although it can better reflect the actual waves and obtain more prepared calculation results, the calculation amount is too large, and a single test project may require several days of calculation time. , which goes against the original intention of rapid and low-cost testing. On the other hand, test items are relatively single and simple. Existing test platforms often only support performance testing of a certain aspect of the unmanned boat, and the testing is independent and single, not holistic. As an autonomous navigation vehicle, unmanned boats need to be tested from hardware foundation to control algorithms, to image recognition, path planning algorithms, etc. If detection is performed in blocks, the integrity of the system is ignored. At the same time, the test items are too simple and cannot be used to test the performance of unmanned boats in various scenarios.

Contents of the invention

In order to overcome the shortcomings and deficiencies of the existing technology, the present invention provides a multi-scenario unmanned boat virtual simulation test platform and a test method. In order to solve the problems existing in the performance testing of existing unmanned boats, the present invention designs a multi-scenario virtual simulation test platform for unmanned boats to realize the integrated testing of unmanned boats in wave environments, multiple scenarios, and multiple projects. Performance Testing. The invention can effectively improve the autonomous navigation performance test results of unmanned boats, reduce test time and reduce test costs.

The present invention adopts the following technical solutions:

A multi-scenario unmanned boat virtual simulation test platform, including a marine environment simulation module, a test evaluation module, a dynamics simulation module, a sensor simulation module and an autonomous navigation interaction module;

The marine environment simulation module is used to combine the Gerstner wave describing the wave shape and the Bretschneider dual parameter spectrum obtained according to the actual wave measurement results to construct a wave scene, and further determine the scene characteristics based on the sea state parameters, test environment parameters and lighting parameters, and generate Ocean test scenario;

The test evaluation module: saves the navigation data of the unmanned boat after the end of each test project, and evaluates the autonomous navigation performance of the unmanned boat according to the evaluation standards of the test project;

The dynamics simulation module includes an unmanned boat planar motion model and a buoyancy calculation model. The motion state of the unmanned boat is obtained through the above models and synchronization with the ocean scene is achieved at the same time;

The autonomous navigation interaction module is used to transmit the sensor signals obtained by the sensor simulation module to the control terminal of the unmanned boat. The control terminal obtains control instructions based on the signals and transmits them back to the unmanned boat virtual test platform in real time to control the virtual unmanned boat. The generated motion of the ocean test scene corresponding to the unmanned boat navigation scene is used to obtain the motion state of the unmanned boat and evaluate it.

Furthermore, the evaluation criteria adopt a hierarchical evaluation.

Furthermore, the hierarchical evaluation is specifically as follows: different test levels are determined according to different sea conditions, different evaluation items are divided into different evaluation elements according to the same test level, and each evaluation in this layer is determined by the analytic hierarchy process. The weight of the evaluation indicators accounted for by the elements is determined by using the cost function method, and the scores are scored based on the ratio of the test data to the ideal data. Finally, the performance of the unmanned boat in this round of testing is evaluated based on the scores.

Furthermore, the planar motion model of the unmanned boat solves and calculates the motion of the unmanned boat on three degrees of freedom in the horizontal plane, and uses the buoyancy calculation model to calculate the motion of the unmanned boat in three degrees of roll, pitch and heave. The motion on the degrees of freedom is calculated to reflect the motion state of the unmanned boat.

Furthermore, the planar motion model of the unmanned boat solves and calculates the motion of the unmanned boat on three degrees of freedom in the horizontal plane. Specifically, based on the MMG control motion equation, the unmanned boat propeller thrust, The influence of wind force, wave force and hydrodynamic force on the motion of unmanned boats in the plane is realized by modifying the coefficients of the MMG steering motion equation for different unmanned boat hulls. The coefficients of the MMG steering motion equation are obtained through numerical simulation.

Furthermore, the buoyancy calculation model calculates the motion of the unmanned boat in three degrees of freedom: roll, pitch and heave, and reflects the motion state of the unmanned boat, specifically as follows:

The unmanned boat model is subdivided into multiple geometric sub-parts along the waterline. Based on the distance from the midpoint of the geometry of each sub-part to the wave surface, the immersed volume of each part is determined. From this, the buoyancy force on each part is calculated. Part of the force acts on the center of gravity of the unmanned boat, causing heaving and rocking motions of the unmanned boat.

Further, the sensor simulation module obtains GPS navigation data, camera, radar data, speed, wind speed, wind direction, acceleration and angular velocity.

Further, the sea state parameters include average wave direction, average wave height, average wave velocity, average wave steepness, average sea breeze direction, and average sea breeze velocity;

The illumination parameters include illumination intensity and water mist concentration.

Further, the unmanned boat was tested on four projects, specifically static obstacle avoidance, dynamic obstacle avoidance, posture maintenance and path tracking.

A testing method for an unmanned boat virtual simulation test platform, including:

Set sea state levels and test items;

Generate ocean test scenarios and related scenarios for testing, and arrange the virtual unmanned boat under test in the scenario;

Sensing information is obtained by virtual sensors and input into the control terminal of the actual unmanned boat;

The control terminal completes the analysis of the sensor data to obtain control instructions, and transmits them back to the virtual test platform to realize navigation control of the virtual unmanned boat;

After the voyage, the voyage data is recorded and evaluation results are generated.

Beneficial effects of the present invention:

The invention realizes the integrated performance test of the unmanned boat in the wave environment in multiple scenarios and multiple projects. The invention can effectively improve the autonomous navigation performance test effect of the unmanned boat, reduce the test time and reduce the test cost.

Description of the drawings

Figure 1 is a general framework diagram of the present invention;

Figure 2 is a flow chart for generating ocean test scenarios by the marine environment simulation module of the present invention;

Figure 3 is a schematic diagram of the virtual unmanned boat and virtual ocean scene of the present invention;

Figure 4 is a schematic flow diagram of the dynamics simulation module of the present invention;

Figure 5 is a schematic flow chart of the test and evaluation module of the present invention;

Figure 6 is a schematic diagram of the virtual simulation test process of the present invention.

Detailed ways

The present invention will be further described in detail below with reference to the examples and drawings, but the implementation of the present invention is not limited thereto.

Example

As shown in Figure 1, a multi-scenario unmanned boat virtual simulation test platform mainly consists of five components, including a marine environment simulation module, a test evaluation module, a dynamics simulation module, a sensor simulation module and an autonomous navigation interaction module. module.

The marine environment simulation module is used to combine the Gerstner wave describing the wave shape and the Bretschneider dual parameter spectrum obtained according to the actual wave measurement results to construct a wave scene, and further determine the scene characteristics based on the sea state parameters, test environment parameters and lighting parameters, and generate Ocean test scenario;

The test evaluation module: saves the navigation data of the unmanned boat after the end of each test project, and evaluates the autonomous navigation performance of the unmanned boat according to the evaluation standards of the test project;

The dynamics simulation module includes an unmanned boat planar motion model and a buoyancy calculation model. The motion state of the unmanned boat is obtained through the above models and synchronization with the ocean scene is achieved at the same time;

The autonomous navigation interaction module is used to transmit the sensor signals obtained by the sensor simulation module to the control terminal of the unmanned boat. The control terminal obtains control instructions based on the signals and transmits them back to the unmanned boat virtual test platform in real time to control the virtual unmanned boat. The generated motion of the ocean test scene corresponding to the unmanned boat navigation scene is used to obtain the motion state of the unmanned boat and evaluate it.

Referring to Figures 2 and 3, a schematic diagram of the generation of the virtual ocean test environment for unmanned boats and a diagram of the virtual unmanned boats and ocean scenes are shown. The sea state parameter setting includes the setting of six parameters, specifically: average wave direction, average wave height, average wave velocity, average wave steepness, average sea breeze direction, and average sea breeze velocity. Lighting parameter settings are used to control the visual field conditions in the scene and test the performance of the unmanned boat under different visibility, including light intensity and water mist concentration. Test environment parameter setting specifically refers to setting the corresponding test items, including the testing of unmanned boats on four items, specifically static obstacle avoidance, dynamic obstacle avoidance, posture maintenance, and path tracking. Through sea state parameter settings, test environment settings and lighting parameter tests, the test scene characteristics are confirmed and the corresponding unmanned boat virtual ocean test environment is generated.

See Figure 4, which shows the schematic diagram of the power simulation module of the unmanned boat virtual simulation platform and the CFD numerical solution calculation diagram of the unmanned boat. The dynamic simulation module includes the unmanned boat planar motion model and buoyancy calculation model. The hull coordinate system is the motion coordinate system of the unmanned boat, represented by the symbol O-xyz. The motion of the unmanned boat can be broken down into six directions, namely surge, heave, heave, roll, pitch and pitch. The motion modeling and calculation under six degrees of freedom are relatively complex, with many undetermined coefficients, and it is difficult to synchronize with the ocean picture. The power calculation of the unmanned boat needs to be simplified accordingly. The motion of the unmanned boat in the plane is calculated and controlled through the planar motion model of the unmanned boat, and the heaving and rocking motion of the unmanned boat is controlled through the buoyancy calculation model. The unmanned boat planar motion model is based on the MMG control motion equation and comprehensively considers the impact of the unmanned boat propeller thrust, wind force, wave force and hydrodynamic force on the unmanned boat's motion in the plane. By modifying the coefficients in the formula, it can adapt to the hulls of different unmanned boats. The coefficients in the formula are obtained through numerical simulation solutions and experimental measurements. Firstly, a CFD three-dimensional numerical calculation pool is established; in view of wind force, wave drift force and hydrodynamic force, several representative groups of working conditions are then selected for numerical calculation and solution; regression analysis and interpolation calculation are performed on the numerical calculation results to determine the corresponding parameters in the mathematical model. coefficient. Regarding the propeller thrust of the unmanned boat, the mooring test method is used to determine the relevant coefficients based on the test results. The specific process of calculating the buoyancy calculation model is as follows: first, subdivide the unmanned boat model into several geometric sub-parts along the water plane. According to the distance from the midpoint of each sub-part geometry to the wave surface, the immersed volume of each part is determined, and the buoyancy force exerted on each part is calculated. Finally, the force of each part is applied to the center of gravity of the unmanned boat, causing heaving and rocking motions of the unmanned boat.

Referring to Figure 5, a schematic diagram of the test and evaluation module of the unmanned boat virtual simulation platform is shown, using a hierarchical evaluation method. First, sea state level matching is performed based on the parameters of the set ocean scene. The schematic diagram shows a total of four levels of sea state classification from level 1 to level 4. The test items are matched according to the set test items, including: static obstacle avoidance, dynamic obstacle avoidance, trajectory tracking and attitude maintenance. Afterwards, the corresponding test elements are determined based on the matching test items. The test elements under both static obstacle avoidance and dynamic obstacle avoidance include: reaction distance, return distance and obstacle avoidance time of the unmanned boat. The test elements under trajectory tracking include: maximum offset degree, average offset degree, and tracking time. The test elements under attitude maintenance include: maximum position deviation, average position deviation, maximum tilt angle, and average tilt angle. Different test levels are determined according to different sea conditions, and different evaluation items are divided into different evaluation elements according to the same test level. The analytic hierarchy process is used to determine the weight of the evaluation indicators occupied by each evaluation element in this layer. The cost function method is used to score based on the ratio of test data to ideal data. Finally, the performance of the unmanned boat in this round of testing is evaluated based on the scores.

See Figure 6, which shows a schematic diagram of the unmanned boat virtual simulation test process. First, set the marine environment through the marine environment module and select the test project. Subsequently, the test ocean environment and test project-related scenes are generated, the virtual unmanned boat under test is arranged in the scene, and the test begins. The tested virtual unmanned boat is modeled in proportion to the actual unmanned boat, and the control circuit and control logic of the virtual unmanned boat should be completely consistent with the actual unmanned boat. The sensor simulation module is equipped with a basic sensor template. Based on the actual sensors on the unmanned boat under test, it simulates the work of various sensors on the unmanned boat by setting the noise function, output information and sensor output power, and releases the sensor information in real time. come out. The above sensor information is sent to the control terminal of the actual unmanned boat in real time through the network serial port. The control terminal of the unmanned boat completes the corresponding data analysis, and then sends the control instructions calculated by the control terminal back to the virtual test platform through the network port, and acts on the control motor of the virtual unmanned boat to realize the control of the virtual unmanned boat. Navigation control. After the voyage is completed, the test evaluation module records the voyage data and generates evaluation results, and the test ends.

The above embodiments are preferred embodiments of the present invention, but the embodiments of the present invention are not limited to the above embodiments. Any other changes, modifications, substitutions, and combinations may be made without departing from the spirit and principles of the present invention. , simplification, should all be equivalent replacement methods, and are all included in the protection scope of the present invention.

Claims (5)

1. A multi-scenario unmanned boat virtual simulation test platform, characterized by including a marine environment simulation module, a test evaluation module, a dynamics simulation module, a sensor simulation module and an autonomous navigation interaction module; The marine environment simulation module is used to combine the Gerstner wave describing the wave shape and the Bretschneider dual parameter spectrum obtained according to the actual wave measurement results to construct a wave scene, and further determine the scene characteristics based on the sea state parameters, test environment parameters and lighting parameters, and generate Ocean test scenario; The test evaluation module: saves the navigation data of the unmanned boat after the end of each test project, and evaluates the autonomous navigation performance of the unmanned boat according to the evaluation standards of the test project; The dynamics simulation module includes an unmanned boat planar motion model and a buoyancy calculation model. The motion state of the unmanned boat is obtained through the above models and synchronization with the ocean scene is achieved at the same time; The autonomous navigation interaction module is used to transmit the sensor signals obtained by the sensor simulation module to the control terminal of the unmanned boat. The control terminal obtains control instructions based on the signals and transmits them back to the unmanned boat virtual test platform in real time to control the virtual unmanned boat. The generated motion of the ocean test scene corresponding to the unmanned boat navigation scene is used to obtain the motion status of the unmanned boat and evaluate it; The evaluation criteria adopt a hierarchical evaluation; The hierarchical evaluation is specifically as follows: different test levels are determined according to different sea conditions, different evaluation items are divided into different evaluation elements according to the same test level, and the analytic hierarchy process is used to determine the requirements of each evaluation element in this layer. The weight of the evaluation index is calculated using the cost function method to score based on the ratio of test data to ideal data. Finally, the performance of the unmanned boat in the same test level is evaluated based on the score; The planar motion model of the unmanned vehicle solves and calculates the motion of the unmanned vehicle in three degrees of freedom in the horizontal plane, and uses the buoyancy calculation model to calculate the three degrees of freedom of the unmanned vehicle in roll, pitch and heave. Calculate the motion on the vehicle to reflect the motion state of the unmanned boat; The planar motion model of the unmanned boat solves and calculates the motion of the unmanned boat on three degrees of freedom in the horizontal plane. Specifically, based on the MMG control motion equation, the unmanned boat propeller thrust, wind force, The influence of wave force and hydrodynamic force on the motion of unmanned boats in the plane is realized by modifying the coefficients of the MMG maneuvering motion equation for different unmanned craft hulls. The coefficients of the MMG maneuvering motion equation are obtained through numerical simulation; The buoyancy calculation model calculates the motion of the unmanned boat in three degrees of freedom: roll, pitch and heave, and reflects the motion state of the unmanned boat, specifically: The unmanned boat model is subdivided into multiple geometric sub-parts along the waterline. Based on the distance from the midpoint of the geometry of each sub-part to the wave surface, the immersed volume of each part is determined, and the buoyancy force on each part is calculated. Part of the force acts on the center of gravity of the unmanned boat, causing heaving and rocking motions of the unmanned boat. 2. The unmanned boat virtual simulation test platform according to claim 1, characterized in that the sensor simulation module obtains GPS navigation data, camera, radar data, speed, wind speed, wind direction, acceleration and angular velocity. 3. The unmanned boat virtual simulation test platform according to any one of claims 1-2, characterized in that the sea state parameters include the average direction of the waves, the average wave height of the waves, the average velocity of the waves, the average steepness of the waves, and the average sea breeze. direction and average sea breeze velocity; The illumination parameters include illumination intensity and water mist concentration. 4. The unmanned boat virtual simulation test platform according to claim 1, characterized in that the unmanned boat is tested on four items, and the four items are specifically static obstacle avoidance, dynamic obstacle avoidance, and posture maintenance. and path tracking. 5. The testing method of the unmanned boat virtual simulation test platform according to any one of claims 1 to 4, characterized in that it includes: Set sea state levels and test items; Generate ocean test scenarios and related scenarios for testing, and arrange the virtual unmanned boat under test in the scenario; Sensing information is obtained by virtual sensors and input into the control terminal of the actual unmanned boat; The control terminal completes the analysis of the sensor data to obtain control instructions, and transmits them back to the virtual test platform to realize navigation control of the virtual unmanned boat; After the voyage, the voyage data is recorded and evaluation results are generated.