Virtual accompanying test system for unmanned surface vehicle

Abstract

The invention discloses a virtual accompanying test system for an unmanned surface vehicle, and relates to the field of unmanned surface vehicle testing, and the system comprises an integrated case, and a task scene simulation module, a sensor simulation module, a pose mapping module, a collision avoidance safety evaluation module and a test display control interface which are integrated in the integrated case. The task scene simulation module constructs a three-dimensional virtual test water area scene based on the static geographic information data, and is provided with a dynamic obstacle and a virtual unmanned ship; the sensor simulation module generates simulation data of the detected unmanned ship environment sensing sensor in the scene; the input end of the pose mapping module is connected with a position and attitude measuring unit which is externally arranged on the case and is arranged on the measured unmanned ship through a data bus, receives real-time position and attitude data of the position and attitude measuring unit, and drives the pose of the virtual unmanned ship to change; the collision avoidance safety evaluation module calculates a total safety evaluation score according to interaction data of the virtual unmanned ship and the dynamic obstacle; and the test display control interface is used for man-machine interaction, test process display and result display.

Description

Technical Field

[0001] This application relates to the field of unmanned surface vessel testing, and in particular to a virtual test system for unmanned surface vessels. Background Technology

[0002] When unmanned surface vessels (USVs) navigate near the coast or in port areas, they often face complex maritime traffic conditions. These areas have high vessel density, numerous offshore structures, and complex traffic flows, which can easily lead to ship collisions.

[0003] Current testing methods for unmanned surface vessels (USVs) collision avoidance safety typically employ model-in-the-loop (MIL) testing, software-in-the-loop (Software-in-the-loop) testing, hardware-in-the-loop (HIL) testing, and sea trials. However, MIL, HIL, and HIL testing often use idealized models to represent the motion control mechanisms of USVs, failing to fully simulate the actual internal mechanisms and components and their operational conditions. Sea trials offer good results and high accuracy, but setting up sea trial scenarios in test ranges is costly, difficult to cover all mission profiles, and involves many uncontrollable factors that can easily affect the navigation of other vessels. Summary of the Invention

[0004] The purpose of this application is to provide a virtual test system for unmanned surface vessels (USVs) that can achieve a comprehensive, efficient, and low-cost evaluation of the collision avoidance performance of USVs.

[0005] To achieve the above objectives, this application provides the following solution: In the first aspect, this application provides a virtual test system for unmanned surface vessels, including an integrated chassis, and a task scenario simulation module, a sensor simulation module, a pose mapping module, a collision avoidance safety assessment module, and a test display and control interface integrated in the integrated chassis; The task scenario simulation module is used to construct a three-dimensional virtual test water area scenario based on static geographic information data, and to set dynamic navigation obstacles and virtual unmanned surface vessels (USVs) in the three-dimensional virtual test water area scenario; the virtual USV is a virtual mapping of the USV under test, and the motion state and sensor data of the virtual USV are fed back and driven to update in real time by the USV under test. The sensor simulation module is used to generate simulation data output from the environmental perception sensors mounted on the unmanned surface vessel under test in the three-dimensional virtual test water scene. The pose mapping module has its input end connected to the position and attitude measurement unit externally placed in the integrated chassis and installed on the unmanned surface vessel under test via a data bus. It is used to receive the real-time position and attitude data of the unmanned surface vessel under test and synchronously drive the virtual unmanned surface vessel in the three-dimensional virtual test water scene to make corresponding pose changes. The collision avoidance safety assessment module is used to calculate the total safety evaluation score based on the interaction data between the virtual unmanned surface vessel and the dynamic obstruction in the three-dimensional virtual test water scene; The test display and control interface is integrated on the all-in-one chassis and is electrically connected to the task scenario simulation module, sensor simulation module and collision avoidance safety assessment module. It is used for human-computer interaction, test process display and result display.

[0006] Optionally, the pose mapping module is connected to the main control computer of the unmanned surface vessel under test through a data output interface, and is used to send the simulation perception data packet generated by the sensor simulation module to the main control computer to drive the decision and control algorithm of the unmanned surface vessel under test.

[0007] Optionally, it also includes a control joystick; the control joystick is connected to the system via an I / O port on the integrated chassis, and is used to receive manual input to control the movement of dynamic navigation obstacles in the three-dimensional virtual test water scene.

[0008] Optionally, the environmental perception sensors simulated by the sensor simulation module include at least one of a camera, marine radar, lidar, and automatic identification system for ships.

[0009] Optionally, the collision avoidance safety assessment module includes: The score calculation unit is used to calculate the score of the evaluation index based on the interaction data between the virtual unmanned surface vessel and the dynamic obstruction in the three-dimensional virtual test water scene; the evaluation index includes collision avoidance success rate, maritime rule violation rate, minimization of impact on the original route, safety distance violation rate, and collision avoidance timing; The weighting determination unit is used to determine the weight of each evaluation indicator based on the score of each evaluation indicator. The overall safety evaluation score calculation unit is used to determine the overall safety evaluation score based on the weighted sum of the weights of each evaluation indicator.

[0010] Optionally, the formula for calculating the collision avoidance success rate is: Q=N 试验次数 / N 总试验次数 ; Where Q is the collision avoidance success rate, and N is the collision avoidance success rate. 试验次数 N represents the number of tests in which a collision occurs. 总试验次数 The total number of trials.

[0011] Optionally, the maritime rule violation rate is determined based on whether the maritime rules are followed during the collision avoidance process and the number of rule violations recorded.

[0012] Optionally, minimizing the impact on the original route includes position oscillation indicators and heading oscillation indicators.

[0013] Optionally, the calculation system for the position oscillation indicator is as follows: Obtain the original route and the actual route; the original route is determined by several points to form a flight segment; the actual route is composed of several GPS coordinate points; Using the vector method, the vertical distance between each GPS coordinate point in the actual route and the original route is determined; The average positional oscillation of the actual route is obtained based on the vertical distance of each GPS coordinate point from the original route. Based on the average positional oscillation of the actual flight path, and according to the positional oscillation scoring rules, the score of the positional oscillation index of the actual flight path is determined.

[0014] Optionally, the calculation system for the heading oscillation index is as follows: Based on the positional relationship between each GPS coordinate point in the actual route and the original route; Determine the heading angle corresponding to each GPS coordinate point in the actual flight route; Calculate the difference in heading angle between adjacent GPS coordinate points along the actual route to obtain the heading angle change sequence; The average value of the heading oscillation is obtained based on the heading angle variation sequence; Based on the average value of the heading oscillation and the heading oscillation scoring rules, the score of the heading oscillation index for the actual route is determined.

[0015] According to the specific embodiments provided in this application, the following technical effects are disclosed: This application provides a virtual test system for unmanned surface vessels (USVs), integrating a mission scenario simulation module, a sensor simulation module, a pose mapping module, a collision avoidance safety assessment module, and a test display and control interface into a single integrated chassis. This reduces the connection complexity and failure risks that may arise from the dispersed arrangement of individual modules. The mission scenario simulation module constructs a three-dimensional virtual test water scene based on static geographic information data and sets dynamic navigation obstacles and virtual USVs. By constructing a three-dimensional virtual scene, the actual test environment can be simulated more realistically, providing near-realistic scenario conditions for USV testing. The sensor simulation module generates simulation data outputs from the environmental perception sensors mounted on the USV under test within the three-dimensional virtual test water scene. This helps simulate the data acquired by the USV's sensors during actual operation in the virtual test environment and can be used to test the correctness and effectiveness of the USV's decision-making and control algorithms based on sensor data. The pose mapping module connects via a data bus to a position and attitude measurement unit externally mounted on the USV within the integrated chassis, receiving real-time position and attitude data of the USV and synchronously driving the virtual USV in the three-dimensional virtual test water scene to make corresponding pose changes. This allows operators to accurately observe and analyze the motion trajectory and attitude changes of the tested unmanned surface vessel (USV) in a virtual environment. The collision avoidance safety assessment module calculates the total safety evaluation score based on the interaction data between the virtual USV and dynamic navigational obstacles in a 3D virtual test water scenario. The test display and control interface is integrated into a single chassis and electrically connected to all major modules, used for human-computer interaction, test process display, and result presentation. This facilitates operator intervention and control of the test process, allowing real-time viewing of various parameters and statuses, and intuitive acquisition of test results. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 A schematic diagram of a virtual test system for unmanned surface vessels provided in an embodiment of this application; Figure 2 A flowchart of a pose mapping module provided in an embodiment of this application; Figure 3 A work cycle diagram of the testing process of the virtual test system for unmanned surface vessels provided in an embodiment of this application; Figure 4 This application provides a schematic diagram of the deployment configuration of a virtual test system for unmanned surface vessels, as shown in one embodiment. Figure 5This is a schematic diagram of the collision avoidance safety assessment module provided in one embodiment of this application. Detailed Implementation

[0018] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0019] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0020] like Figure 1 As shown, this embodiment provides a virtual test system for unmanned surface vessels, including an integrated chassis, and a task scenario simulation module, a sensor simulation module, a pose mapping module, a collision avoidance safety assessment module, and a test display and control interface integrated in the integrated chassis; The task scenario simulation module is used to construct a three-dimensional virtual test water area scenario based on static geographic information data, and to set dynamic navigation obstacles and virtual unmanned surface vessels (USVs) in the three-dimensional virtual test water area scenario; the virtual USV is a virtual mapping of the USV under test, and the motion state and sensor data of the virtual USV are fed back and driven to update in real time by the USV under test. The sensor simulation module is used to generate simulation data output from the environmental perception sensors mounted on the unmanned surface vessel under test in the three-dimensional virtual test water scene. The pose mapping module has its input end connected to the position and attitude measurement unit externally placed in the integrated chassis and installed on the unmanned surface vessel under test via a data bus. It is used to receive the real-time position and attitude data of the unmanned surface vessel under test and synchronously drive the virtual unmanned surface vessel in the three-dimensional virtual test water scene to make corresponding pose changes. The collision avoidance safety assessment module is used to calculate the total safety evaluation score based on the interaction data between the virtual unmanned surface vessel and the dynamic obstruction in the three-dimensional virtual test water scene; The test display and control interface is integrated on the all-in-one chassis and is electrically connected to the task scenario simulation module, sensor simulation module and collision avoidance safety assessment module. It is used for human-computer interaction, test process display and result display.

[0021] Specifically, the system uses the robot simulation software Webots and the Robot Operating System (ROS) as core components, employing digital twin and mixed reality technologies to address the low cost-effectiveness of mission scenario construction during unmanned surface vessels (USVs) real-world testing. Targeting the typical USV operational paradigm of "situational awareness – decision-making and planning – motion control," this product replaces the "situational awareness" component, directly simulating and generating the results in virtual space and injecting them via bus into the perception data receiving port of "decision-making and planning – motion control." This blends virtual "situational awareness" with real "decision-making and planning – motion control," achieving coupled verification and testing of the two key technologies of USVs: "decision-making and planning – motion control."

[0022] like Figure 1 As shown, a virtual test system for unmanned surface vessels includes a GNSS antenna, a position and attitude measurement unit, a test display and control interface, a keyboard and mouse, an integrated chassis, a quick display and control panel, I / O ports, and control joysticks, among which: The GNSS antenna needs to be installed and fixed to the hull of the unmanned surface vessel according to the instruction manual, in order to receive satellite positioning data; The position and attitude measurement unit needs to be installed and fixed by the user amidships of the unmanned surface vessel (USV) according to the instruction manual. It measures the USV's position and attitude in the Earth coordinate system in real time by acquiring data such as acceleration, angular velocity, and heading angle, as well as satellite positioning data received by the GNSS antenna. After installation, the position and attitude measurement unit needs to be calibrated at sea. The user should then pilot the USV out to sea for a few nautical miles, and the automatic calibration can be completed by combining the position and attitude measurement unit. The test display and control interface is the main human-computer interaction interface of the unmanned surface vessel mission profile injector. It includes test case editing and loading, test parameter setting, dynamic display of the test process, and test data recording and export function modules. Test personnel can operate the mission profile injector in a variety of ways, such as touch, keyboard and mouse operation, and shortcut key switches. The keyboard and mouse are used to operate the experimental display and control interface; The quick control panel is used to operate the test display and control interface. It has several commonly used commands preset and several user-defined commands reserved. The I / O port is used to import test cases and export test data, and can also be used to connect external operation joysticks.

[0023] The joystick is used to control dynamic scene elements in the virtual space, such as the test boat and the mother ship, which are manually operated.

[0024] In this embodiment, the task scenario simulation module process includes two parts. The first is scenario simulation based on static geographic information. First, the latitude and longitude range, geographic coordinate system, and projected coordinate system range of the determined test area are obtained. The area within this latitude and longitude range is selected from an open-source map simulation website and exported as a map file containing the specific shape information of landmark buildings, ground information, road network information, and the location and planar dimensions of buildings. The map file is processed using modeling tools, performing coordinate transformations (latitude and longitude to local coordinate system) according to the geographic coordinate system and projected coordinate system, converting planar buildings into three-dimensional building entities, creating ground thickness, adding water bodies, performing Boolean operations with the land area, preserving the water body range, performing material rendering, and repairing error models. The file is then saved as an .obj file. To improve scene running efficiency, rendering can be omitted, and the file can be saved as an .stl file. The .obj file is imported into the task scenario simulation module, and the simulation module displays the above scenario. The second part is navigation obstacle simulation, which adds the target ship and buoy obstacles required for the test to the static geographic information scene. There are two ways to place obstructions: one is to use actual data to obtain waypoint information from the AIS data of all ships in a certain time and space in the current area, assign the waypoint latitude and longitude to the target ship, and assign speed information; the other is to artificially design extreme working conditions or edge scenarios, plan the position of obstructions in a static geographic information scene, and assign speed information.

[0025] In this embodiment, the sensor simulation module process is as follows: Different unmanned surface vessels (USVs) under test may be equipped with different environmental perception sensors. The virtual test system for USVs includes common USV sensors such as cameras, marine radar, lidar, and AIS. The camera model uses a frustum model to define what is visible from the camera. It does not render any 3D shapes completely outside this frustum; shapes too close to the camera (between the camera and the near-plane) will not be displayed. Camera simulation model parameters include resolution, field of view, and noise level. The camera simulation output is a task simulation scene image.

[0026] The main calculation formulas for the radar model are as follows: radar range equation, and received power Pr determines whether the target is detectable.

[0027] ; In the formula, P t For transmission power, G t G r Let P be the transmit / receive antenna gain, λ be the wavelength, σ be the target radar cross-section, R be the target range, and L be the system loss. r The target is detected when the receiver sensitivity is ≥. Other radar model parameters include minimum range, maximum range, and field of view. The radar simulation and mission scenario simulation interact to output a target list, including information such as position, velocity, and heading.

[0028] The LiDAR model emits multiple virtual rays, each emanating from an origin (the sensor's location) along a specific directional vector. The goal is to find the intersection point of this ray with the nearest object in the scene (an object with a collision bounding box).

[0029] The AIS model performs coordinate system transformation on the position of the virtual obstructing vessels in the simulation environment and outputs latitude and longitude information in WGS84. The model outputs the information in plain AIS code, including the number of AIS targets, target MMSI number, target IMO number, target latitude and longitude, heading, etc.

[0030] The pose mapping module is connected to the main control computer of the unmanned surface vessel under test through a data output interface. It is used to send the simulation perception data packets generated by the sensor simulation module to the main control computer to drive the decision and control algorithms of the unmanned surface vessel under test.

[0031] Specifically, such as Figure 2 As shown, the virtual test system acquires the real-time position and attitude of the actual unmanned surface vessel (USV) during its navigation and synchronously maps it onto the USV under test in the virtual environment. The position mapping includes both position mapping and attitude mapping. The position mapping process is shown in the figure below. First, the coordinate system drift caused by the simulation coordinate transformation and the difference between the real coordinate system is calculated. Then, the position of the vessel in the simulation environment is calculated by subtracting the coordinate system offset from the obtained latitude and longitude position of the actual vessel.

[0032] Attitude mapping simulates the unmanned surface vessel under test to directly obtain heading, roll, pitch, and yaw information from the navigation system of the actual unmanned surface vessel and to respond in real time to attitude changes.

[0033] In this embodiment, the collision avoidance safety assessment module assessment process is as follows: Figure 5 As shown, to evaluate safety, the scores for the secondary indicators of collision avoidance success rate, maritime rule violation rate, minimization of impact on the original route, safety distance violation rate, and collision avoidance timing need to be calculated separately, and each secondary indicator is assigned a weight, and the total safety evaluation score is obtained by weighted summation.

[0034] Specifically, the formula for calculating the collision avoidance success rate is as follows: Q=N 试验次数 / N 总试验次数 ; Where Q is the collision avoidance success rate, and N is the collision avoidance success rate. 试验次数 N represents the number of tests in which a collision occurs. 总试验次数 The total number of trials.

[0035] In the experiment, the distances {d1,d2,…,d} between the ship and the virtual obstacle were obtained in real time. n}, The list is updated every 10s. Compare the distance with the sum of the collision radius r1 of the ship itself and the collision radius of the target ship, i.e., r1 + r2. If d n < r1 + r2, it is recorded as a collision. The collision avoidance success rate score is determined by the result itself. The 90% collision avoidance success rate score is 100.

[0036] The maritime rule violation rate determines whether the collision avoidance process complies with maritime rules through manual judgment and records the number of rule violations. Accumulate the values on the evaluation interface to assist in counting. The maritime rule violation rate (only for scenarios with obstacle ships), the number of times of violating collision avoidance rules during collision avoidance / the number of collision avoidance times (which can be understood as the number of obstacle ships), a 0% maritime rule violation rate is recorded as 100.

[0037] Calculating the position oscillation and course oscillation to minimize the impact on the original route. Position oscillation: The original route is a section (line segment) determined by several points. The actual route consists of several GPS coordinate points. Calculate the perpendicular distance from the coordinate points to the line segment, and use vectors to solve. The position relationship between the point and the line segment is that the point is inside the line segment. Constrain the calculation of GPS coordinate points and the current section through the position constraint relationship between the point and the line segment (as shown in the figure). Finally, calculate the average value of the oscillation. Course oscillation: By judging the position relationship between GPS waypoints and the line segment, determine which points belong to the current section. Obtain the course information of the current section and the course information of the GPS waypoints belonging to the current section. Calculate the difference between the course information of the current section and the course of the GPS waypoints to obtain the average value of the course oscillation. There is a mapping relationship between the degree of impact and the score. To minimize the impact on the original route, if the average value of position oscillation < 1 nautical mile, it is recorded as 100; if the average value of position oscillation > 1 nautical mile and < 2 nautical miles, it is recorded as 90, and so on. If the average value of course oscillation < 10°, it is recorded as 100; if the average value of course oscillation > 10° and < 20°, it is recorded as 90, and so on.

[0038] Minimize the impact on the original route: According to the above determination methods of position oscillation and course oscillation and the corresponding mapping relationship between the degree of impact and the score, try to make the average values of position oscillation and course oscillation in a lower range as much as possible to achieve the purpose of minimizing the impact on the original route.

[0039] The calculation method of the safety distance violation rate is the same as that of the collision avoidance success rate, but the collision radius becomes the safety radius. If the number of times of not maintaining the safety distance is 0, it is recorded as 100 points; if the number of times of not maintaining the safety distance is greater than 0 and less than 10, it is recorded as 90 points. And so on.

[0040] The timing of collision avoidance is determined by obtaining the flags of the actual ship's algorithm information. Simultaneously, the distance between the ship and other ships at the current moment is recorded. If the distance is less than the safe radius, the collision avoidance timing is considered unreasonable. If every collision avoidance timing is maintained outside the safe distance, it is scored as 100 points; if 90% of the collision avoidance timings in a test are reasonable, it is scored as 90 points, and so on.

[0041] In this embodiment, the working cycle of the unmanned surface vessel virtual test system is as follows: Figure 3 As shown: The first step is to solve the virtual scene at time T in the virtual space (task profile injector) and obtain situational awareness information such as the relative position, relative motion direction, and relative motion speed of the moving and static targets around the unmanned vessel twin and the test unmanned vessel, and package them in the form of a list. The second step is that in the real space (actual test site), after the unmanned surface vessel under test receives situational awareness information from the surrounding twin in the virtual space through the bus interface, the "decision planning - motion control" algorithm on the test object autonomously controls the movement of the test object in the real sea area. The third step is to obtain the position and attitude of the unmanned surface vessel at time T+1 in real space (actual test site). There are two methods: one is to install a combined inertial measurement unit on the unmanned surface vessel, and the other is to obtain the position and attitude directly from the end system of the vessel. The fourth step is to synchronously map the position and attitude data of the unmanned surface vessel at time T+1 in the virtual space (mission profile injector) to the position and attitude of the twin in the virtual space. Fifth, repeat the above steps.

[0042] The system deployment form is as follows Figure 4 As shown, the first step is to install and fix the integrated inertial navigation measurement unit (INS) at a suitable location on the actual unmanned surface vessel (USV). Then, the INS is connected to the USV mission profile injector via a data bus. Finally, the USV mission profile injector is connected to the situational awareness interface of the USV's main control computer (client equipment) via another data bus to begin testing the USV's decision-making, planning, and control algorithms. During testing, personnel can use a joystick to control a virtual test vessel (or mother ship) to coordinate with the USV under test, or act as the "blue force" to disrupt the USV's mission execution, achieving both "human-machine collaboration" and "human-machine confrontation."

[0043] In summary, this application has the following technical effects: This application proposes a virtual test system for autonomous missions of unmanned surface vessels (USVs). It can effectively map real USV intelligent algorithms to a high-fidelity environment simulated by Webots' physics simulation engine through ROS middleware. Furthermore, it can achieve more realistic algorithm and operating condition testing through perception and interaction with simulated sensors in the simulation environment, effectively reducing the risks of physical testing, lowering testing costs, improving the coverage of test conditions, and shortening the development cycle.

[0044] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0045] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. In summary, the content of this specification should not be construed as a limitation of this application.

Claims

1. A virtual test system for unmanned surface vessels, characterized in that, It includes an integrated chassis, and a task scenario simulation module, a sensor simulation module, a pose mapping module, a collision avoidance safety assessment module, and a test display and control interface integrated in the integrated chassis; The task scenario simulation module is used to construct a three-dimensional virtual test water area scenario based on static geographic information data, and to set dynamic navigation obstacles and virtual unmanned surface vessels (USVs) in the three-dimensional virtual test water area scenario; the virtual USV is a virtual mapping of the USV under test, and the motion state and sensor data of the virtual USV are fed back and driven to update in real time by the USV under test. The sensor simulation module is used to generate simulation data output from the environmental perception sensors mounted on the unmanned surface vessel under test in the three-dimensional virtual test water scene. The pose mapping module has its input end connected to the position and attitude measurement unit externally placed in the integrated chassis and installed on the unmanned surface vessel under test via a data bus. It is used to receive the real-time position and attitude data of the unmanned surface vessel under test and synchronously drive the virtual unmanned surface vessel in the three-dimensional virtual test water scene to make corresponding pose changes. The collision avoidance safety assessment module is used to calculate the total safety evaluation score based on the interaction data between the virtual unmanned surface vessel and the dynamic obstruction in the three-dimensional virtual test water scene; The test display and control interface is integrated on the all-in-one chassis and is electrically connected to the task scenario simulation module, sensor simulation module and collision avoidance safety assessment module. It is used for human-computer interaction, test process display and result display.

2. The virtual test system for unmanned surface vessels according to claim 1, characterized in that, The pose mapping module is connected to the main control computer of the unmanned surface vessel under test through a data output interface. It is used to send the simulation perception data packets generated by the sensor simulation module to the main control computer to drive the decision and control algorithm of the unmanned surface vessel under test.

3. The virtual test system for unmanned surface vessels according to claim 2, characterized in that, It also includes a joystick; the joystick is connected to the system via an I / O port on the integrated chassis and is used to receive manual input to control the movement of dynamic navigation obstacles in the three-dimensional virtual test water scene.

4. The virtual test system for unmanned surface vessels according to claim 1, characterized in that, The sensor simulation module simulates environmental perception sensors including at least one of cameras, marine radar, lidar, and automatic identification systems for ships.

5. The virtual test system for unmanned surface vessels according to claim 1, characterized in that, The collision avoidance safety assessment module includes: The score calculation unit is used to calculate the score of the evaluation index based on the interaction data between the virtual unmanned surface vessel and the dynamic obstruction in the three-dimensional virtual test water scene; the evaluation index includes collision avoidance success rate, maritime rule violation rate, minimization of impact on the original route, safety distance violation rate, and collision avoidance timing; The weighting determination unit is used to determine the weight of each evaluation indicator based on the score of each evaluation indicator. The overall safety evaluation score calculation unit is used to determine the overall safety evaluation score based on the weighted sum of the weights of each evaluation indicator.

6. The virtual test system for unmanned surface vessels according to claim 5, characterized in that, The formula for calculating the collision avoidance success rate is: Q=N 试验次数 / N 总试验次数 ; Where Q is the collision avoidance success rate, and N is the collision avoidance success rate. 试验次数 N represents the number of tests in which a collision occurs. 总试验次数 The total number of trials.

7. The virtual test system for unmanned surface vessels according to claim 6, characterized in that, The maritime rule violation rate is determined based on whether the maritime rules are followed during collision avoidance and by recording the number of rule violations.

8. The virtual test system for unmanned surface vessels according to claim 7, characterized in that, Minimizing the impact on the original route includes position oscillation indicators and heading oscillation indicators.

9. The virtual test system for unmanned surface vessels according to claim 8, characterized in that, The calculation system for the position oscillation indicator is as follows: Obtain the original route and the actual route; the original route is determined by several points to form a flight segment; the actual route is composed of several GPS coordinate points; Using the vector method, the vertical distance between each GPS coordinate point in the actual route and the original route is determined; The average positional oscillation of the actual route is obtained based on the vertical distance of each GPS coordinate point from the original route. Based on the average positional oscillation of the actual flight path, and according to the positional oscillation scoring rules, the score of the positional oscillation index of the actual flight path is determined.

10. A virtual test system for unmanned surface vessels according to claim 8, characterized in that, The calculation system for the heading oscillation index is as follows: Based on the positional relationship between each GPS coordinate point in the actual route and the original route; Determine the heading angle corresponding to each GPS coordinate point in the actual flight route; Calculate the difference in heading angle between adjacent GPS coordinate points along the actual route to obtain the heading angle change sequence; The average value of the heading oscillation is obtained based on the heading angle variation sequence; Based on the average value of the heading oscillation and the heading oscillation scoring rules, the score of the heading oscillation index for the actual route is determined.

Images