Port automatic driving vehicle simulation test method

Abstract

The invention relates to the technical field of automatic driving vehicle testing, and particularly discloses a port automatic driving vehicle simulation testing method, which comprises the following steps: step 1, analyzing testing requirements, step 2, modeling a port and a vehicle, step 3, preparing for testing, step 4, selecting testing types, step 5, determining testing items, step 6, selecting testing working conditions, and step 7, executing simulation testing. And 8, analyzing and evaluating a result. The testing method has the advantages that three-level classification is carried out based on the testing types, the testing items and the testing working conditions, the macroscopic testing types, the mesoscopic testing items and the microscopic testing working conditions are progressive layer by layer and are strict in logic, the design process of a testing scene is clear, and the testing efficiency is improved. And meanwhile, summarization, analysis and optimization of subsequent test results are facilitated, the overall efficiency and quality of test work are improved, and an efficient and accurate path is developed for verification and evaluation of the port automatic driving technology.

Description

Technical Field

[0001] The present invention relates to the technical field of autonomous driving vehicle testing, and in particular to a simulation testing method for an autonomous port vehicle. Background Art

[0002] In recent years, the Ministry of Transport has vigorously promoted port automation and intelligence, and my country's port automation level has been at the forefront internationally. The unique traffic environment of the port makes it an ideal testing ground for autonomous driving, and the prominent position of Chinese ports in global cargo throughput has stimulated a strong demand for unmanned driving technology. In order to further promote the implementation of port autonomous driving technology, in addition to firm support at the policy level and improved infrastructure construction, strengthening large-scale testing of autonomous driving container trucks is a crucial link. This step is not only a direct verification of the maturity and reliability of the technology, but also a key turning point in promoting autonomous driving technology from theory to practical application, and from small-scale pilots to large-scale deployment. There are two main aspects to strengthening large-scale testing of autonomous driving container trucks: building diversified test scenarios and increasing the scale and frequency of tests.

[0003] At present, autonomous driving in ports is in a booming stage. The testing field of autonomous container trucks generally focuses on field testing, but it is often limited by the busy operating environment of the port, resulting in bottlenecks in time and space. In order to break through this limitation, simulation testing technology came into being. Its advantage is that it can conduct tests flexibly and efficiently without interfering with the daily operations of the port, thereby effectively avoiding the space and time constraints of field testing. However, despite the huge potential of simulation testing, its application is still insufficient. Existing simulation tests often focus on a few types of preset scenarios, resulting in fragmentation of simulation test scenarios, lack of coherence and logic of simulation test scenarios, and narrow coverage of simulation test scenarios. A systematic and comprehensive testing system has not yet been established. Summary of the invention

[0004] The present invention proposes a port autonomous driving vehicle simulation test method in order to solve the above problems.

[0005] The technical solution of the present invention is achieved in this way:

[0006] A port autonomous driving vehicle simulation test method comprises the following steps:

[0007] Step 1: Analyze the test requirements. Analyze the test requirements according to the test procedures for the automatic driving function of intelligent connected vehicles, determine the type of test, the test items to be selected, and the test conditions;

[0008] Step 2: Port and vehicle modeling: Establish a simulation model based on a typical port layout. The port model includes berths, container yards, warehouses, bridges, loading and unloading equipment, traffic signs, and the road network connecting them. Select the parameters of the test vehicle according to the needs, including vehicle size, vehicle weight, vehicle power, speed, acceleration, braking data, and power loss ratio, and import the port autonomous driving vehicle model to be tested;

[0009] Step 3: Test preparation, loading simulation software modules, including scheduling task module, path planning module, data perception module, and decision-making module;

[0010] Step 4: Test type selection. Test types include functional testing, performance testing, safety testing, reliability testing, and boundary testing. Select the test type according to the test requirements.

[0011] Step 5: Determine the test items and select specific test items according to the test type;

[0012] Step 6: Select the test condition. Select the test condition according to the test type and test items, including automatic lane change test condition, emergency braking test condition, and extreme weather test condition;

[0013] Step 7: Execute simulation tests, and conduct tests in sequence according to the selected test type, test items, and test conditions, and collect the operating data of the unmanned vehicle during the test, including the vehicle's position, speed, acceleration, sensor data, and decision instructions;

[0014] Step 8: Result analysis and evaluation: Analyze the collected data, compare the test indicators and acceptance criteria, determine the performance of the unmanned vehicle in each test, and generate a detailed test report to provide a basis for the actual application of autonomous vehicles in ports.

[0015] Furthermore, in step 5, each test type includes at least two test items, and each test item has a test indicator and an acceptance criterion.

[0016] Furthermore, in step 5, 1 to 3 test items may be selected simultaneously for composite testing.

[0017] Furthermore, in step 6, 1 to 3 test conditions may be selected simultaneously for composite testing.

[0018] Furthermore, the content of the test report in step 8 includes a description of the test process, an analysis of the test results, existing problems and improvement suggestions.

[0019] By adopting the above technical scheme, the beneficial effects of the present invention are as follows: the testing method of the present invention is classified into three levels based on test type, test items and test conditions, from macro test type to meso test items and then to micro test conditions, which is progressive and logically rigorous, not only making the design process of the test scenario clear and organized, but also facilitating the summary, analysis and optimization of subsequent test results, improving the overall efficiency and quality of the testing work, and opening up an efficient and accurate path for the verification and evaluation of port autonomous driving technology; the testing method of the present invention can systematically cover various complex situations and extreme challenges that port autonomous driving technology may encounter, ensuring the comprehensiveness and diversity of the test scenarios, effectively avoiding test blind spots, and providing a more solid guarantee for the reliability and safety of autonomous driving technology; high-coverage test scenarios not only help to discover potential defects and deficiencies in the technology, but also provide valuable data support for the continuous improvement and optimization of the technology, greatly improving the scientificity and practicality of the test scenario construction. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0021] Figure 1 It is a flow chart of the present invention. DETAILED DESCRIPTION

[0022] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. 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.

[0023] like Figure 1 As shown, a port autonomous driving vehicle simulation test method includes the following steps:

[0024] Step 1: Analyze the test requirements. Analyze the test requirements according to the test procedures for the automatic driving function of intelligent connected vehicles, determine the type of test, the test items to be selected, and the test conditions;

[0025] Step 2: Port and vehicle modeling: Establish a simulation model based on a typical port layout. The port model includes berths, container yards, warehouses, bridges, loading and unloading equipment, traffic signs, and the road network connecting them. Select the parameters of the test vehicle according to the needs, including vehicle size, vehicle weight, vehicle power, speed, acceleration, braking data, and power loss ratio, and import the port autonomous driving vehicle model to be tested;

[0026] Step 3: Test preparation, loading simulation software modules, including scheduling task module, path planning module, data perception module, and decision-making module;

[0027] Step 4: Test type selection. Test types include functional testing, performance testing, safety testing, reliability testing, and boundary testing. Select the test type according to the test requirements.

[0028] Functional testing: for the basic driving functions of autonomous vehicles in the port, such as lane keeping, automatic lane change, adaptive cruise control, automatic loading and unloading control functions, etc. Simulate the preset path in the port map to check whether the vehicle can accurately drive along the planned route and accurately perform loading and unloading operations in the loading and unloading area. These test items not only ensure the usability of the system in daily driving, but also lay a solid foundation for subsequent higher-level tests.

[0029] Performance testing includes driving performance testing, perception performance testing, and decision-making performance testing:

[0030] Driving performance test: evaluates the vehicle's speed control accuracy, acceleration stability, braking performance, etc. Under different load conditions (simulating cargo weight), the vehicle's driving parameters are measured in straight-line driving, turning, uphill and downhill conditions, and compared with the preset performance indicators.

[0031] Perception performance test: examines the vehicle's sensors (lidar, camera, millimeter-wave radar, etc.) in terms of their ability to detect obstacles (such as other vehicles, cargo piles, terminal facilities, etc.) in the port environment, as well as their detection range, detection accuracy, and adaptability to environmental changes (weather, lighting conditions).

[0032] Decision-making performance test: Analyze the rationality of vehicle decisions in complex traffic scenarios (intersections, multiple vehicles operating simultaneously, etc.), such as selecting the optimal driving path, the correctness of avoidance strategies, etc.

[0033] Reliability test: Conduct long-term simulation runs to simulate the situation where the vehicle operates continuously at the port for several days or even weeks, check whether the software system has any faults (such as program crashes, memory leaks, etc.), whether the hardware system has any abnormal conditions such as overheating and component damage, and evaluate the degree of performance degradation of the vehicle after long-term operation.

[0034] Safety testing is an assessment of the autonomous driving system's ability to respond to potentially dangerous scenarios. By simulating scenarios such as collision warning and emergency braking, the system is tested to see whether it can identify and respond to dangers in a timely and accurate manner, thereby ensuring driving safety. This includes collision safety testing and information security testing:

[0035] Collision safety test: simulates scenarios where a vehicle collides with obstacles of different types and speeds, checks the effectiveness of the vehicle's passive safety systems (such as seat belts, anti-collision structures, etc.), and whether the autonomous driving system has taken reasonable avoidance or braking measures before the collision.

[0036] Information security testing: Detect whether the vehicle's communication system and control system have vulnerabilities that can be attacked by hackers, simulate malicious attack scenarios, such as tampering with driving instructions, interfering with sensor data, etc., to evaluate the vehicle's information security protection capabilities.

[0037] Boundary testing: Further challenges the limits of the system, and verifies the adaptability and stability of the system in different environments through tests of boundary conditions such as extreme weather, night driving, and complex traffic environments. This type of test helps to discover potential defects and deficiencies in the system, and provides direction for subsequent optimization and improvement.

[0038] Step 5: Determine the test items. Select specific test items based on the test type. Each test type includes at least two test items. These items are not only the concretization and instantiation of the test type, but also the in-depth verification of the various capabilities of the autonomous driving system.

[0039] The functional test items include: path planning accuracy test items, automatic lane change function test items, loading and unloading action accuracy test items, vehicle and port equipment interaction function test items, etc.

[0040] Performance test items include: speed stability test items, sensor accuracy test items, complex scenario decision time test items, etc.

[0041] Reliability test items include: system long-term operation stability test items, hardware durability test items, etc.

[0042] Safety test items include: simulation test items of different collision scenarios, emergency braking test items at different vehicle speeds, network attack protection capability test items, etc.

[0043] Boundary test projects include: different extreme weather scenario test projects, different visibility scenario test projects, multi-party participation chaotic traffic flow test projects, etc.

[0044] Each test item determines detailed test indicators and acceptance criteria. For example, in the path planning accuracy test item, the acceptance criterion may be that the deviation between the actual driving path of the vehicle and the planned path is within a certain range (such as less than [X] centimeters). In the sensor accuracy test item, for lidar, the distance error of the detected obstacle should be less than [specific value], etc.

[0045] Step 6: Select the test condition. Select the test condition according to the test type and test items, including automatic lane change test condition, emergency braking test condition, and extreme weather test condition;

[0046] Automatic lane change test conditions: The vehicles in front of the autonomous vehicle and in the lane to the left are driving slowly, simulating overtaking scenarios in the real world. This realistic simulation helps verify the system's lane change decision-making and execution capabilities during actual overtaking.

[0047] Emergency braking test condition: By simulating an emergency situation such as a pedestrian crossing the road behind an obstacle on the right side of the autonomous vehicle, the system's response speed and handling ability to emergency events are tested. This test condition not only tests the system's perception and decision-making capabilities, but also verifies the performance of its emergency braking system.

[0048] Extreme weather test conditions: By simulating driving scenarios under extreme weather conditions such as rain, snow, fog and haze, the system's adaptability and stability under different weather conditions can be verified. Such test conditions help to discover potential problems of the system in different environments and provide direction for its subsequent optimization and improvement.

[0049] The automatic lane change test condition and the emergency brake test condition are both traffic flow test conditions. According to the actual operation of the port, the flow, driving speed and driving route of different types of vehicles (container trucks, forklifts, automatic guided vehicles, etc.) are set. The starting, stopping, acceleration, deceleration and other behaviors of the vehicle, as well as the interaction between vehicles (overtaking, giving way, etc.) are simulated. Different traffic flow density scenarios are set considering the peak and off-peak hours of port operations.

[0050] The extreme weather test conditions are simulated through weather condition simulation: including different weather conditions such as sunny days, rainy days, foggy days, snowy days, etc., simulating the impact of weather on sensors (such as camera visibility, lidar reflectivity, etc.); lighting condition simulation: considering the day, night, sunlight at different angles and the lighting conditions of port lighting facilities, analyzing the effect of lighting on the vehicle perception system; other environmental factors: such as temperature, humidity, etc., which have an impact on vehicle hardware performance, are also included in the simulation range.

[0051] Step 7: Execute simulation tests, and conduct tests in sequence according to the selected test type, test items, and test conditions, and collect the operating data of the unmanned vehicle during the test, including the vehicle's position, speed, acceleration, sensor data, and decision instructions;

[0052] Step 8: Result analysis and evaluation: analyze the collected data, compare the test indicators and acceptance criteria, determine the performance of the unmanned vehicle in each test, and generate a detailed test report. The content of the test report includes a description of the test process, analysis of the test results, existing problems and improvement suggestions, providing a basis for the actual application of autonomous driving vehicles in ports.

[0053] Examples of test types and test items in the test method of the present invention are as follows:

[0054] ①. Test type: recognition and response to signs and markings; Test items: recognition and response to common signs and markings in ports.

[0055] This test only targets common signs and markings in the port area. Non-existent or uncommon markings and signs in the port area are not included in this test. It verifies whether the autonomous vehicle can accurately identify and respond to common traffic signs and markings in the port, and requires the autonomous vehicle to take appropriate measures such as slowing down, stopping or detouring when approaching these signs.

[0056] ②. Test type: obstacle recognition and response; Test items: Test the vehicle's ability to recognize stationary obstacles and make corresponding braking, avoidance and other operations.

[0057] Restrictions on obstacle types have been added. Only static obstacles are tested, and dynamic obstacles are not within the scope of this test. It is required that autonomous vehicles should be able to accurately identify these obstacles within a safe distance and take corresponding braking, deceleration or avoidance measures to ensure driving safety.

[0058] ③. Test type: severe weather identification. Test items: Test the vehicle’s identification capability under the influence of severe weather.

[0059] The project tests the perception capabilities of autonomous vehicles in bad weather conditions with limited vision or slippery roads. Autonomous vehicles should be able to continuously and stably identify road signs, lane lines, surrounding vehicles and pedestrians to maintain safe driving.

[0060] ④. Test type: Identification and response to the driving status of the vehicle ahead. Test items: Timely identification of acceleration, deceleration, braking, etc. of the vehicle ahead and corresponding response.

[0061] The driver's ability to recognize and respond to these changes in real time is tested by the possible behavior of the vehicle in front and the anticipated capabilities of the autonomous vehicle. The autonomous vehicle needs to adjust its speed, lane position or take evasive measures based on the behavior of the vehicle in front.

[0062] ⑤. Test type: encountering a vehicle coming from the side when leaving the work area. Test item: encountering a vehicle coming from the side when starting to leave the work area after completing loading and unloading operations.

[0063] Based on its current location, the target vehicle's initial location and its direction of movement, the autonomous vehicle needs to observe and predict the direction, speed and distance of the oncoming vehicle in advance, and take necessary avoidance measures to ensure safe exit.

[0064] ⑥. Test type: Pedestrian recognition and avoidance; Test items: Recognition and response to staff members in different directions in the port.

[0065] By designing test scenarios for pedestrians walking in different directions, and anticipating that autonomous driving vehicles can accurately identify pedestrians walking in different directions, slow down in advance, observe pedestrian movements and take avoidance measures to protect pedestrian safety.

[0066] ⑦. Test type: Turn left into the gantry crane box area and encounter the gantry crane box exit area. Test item: The test vehicle is driving on the main road and is about to turn left into the box area when encountering the gantry crane box exit area.

[0067] By refining the position and driving intention of the autonomous vehicle at the beginning of the test, and clarifying the corresponding behavior of the target gantry crane at the specific moment when the autonomous vehicle turns left, it is verified whether the autonomous vehicle can correctly identify the moving gantry crane while turning and respond.

[0068] ⑧. Test type: roadside parking. Test items: testing the vehicle’s parking capacity in different parking spaces and its ability to start and leave the parking space.

[0069] It is specified that the roadside parking required in the test type needs to be parked in a fixed parking space, and the autonomous vehicle needs to park smoothly, not cross the line, and not collide with surrounding obstacles. In addition to testing the parking ability of the autonomous vehicle, the ability of the autonomous vehicle to start and leave the parking space also needs to be tested.

[0070] ⑨. Test type: precise positioning under the crane. Test items: testing the longitudinal relative positioning capability of the automatic driving system and the crane.

[0071] Since the self-driving container trucks operating in the port need to correctly park at the top parking space under the crane to complete cargo loading and unloading, this test project requires testing whether the self-driving vehicle algorithm can accurately identify the parking space under the crane and complete the parking action at the corresponding position.

[0072] Through further analysis of the test items, the most specific and detailed hierarchical test conditions are obtained. The test conditions specify the target type, test location, and the behavior of the host vehicle and the target vehicle to ensure the comprehensiveness and effectiveness of the test. For example:

[0073] (1) Test items: Recognition and response to common port signs and markings.

[0074] The test conditions under this test item include recognition and response to common port signs and markings such as stop signs (signs), traffic lights, speed limit signs, construction signs, etc.

[0075] Test items: Test the vehicle's ability to identify stationary obstacles and make corresponding braking, avoidance and other operations;

[0076] This test condition specifies the types of stationary obstacles, including stationary vehicles ahead, construction sections ahead, and stationary obstacles ahead.

[0077] (2) Test items: Test the vehicle’s recognition capability under the influence of adverse weather conditions.

[0078] The test conditions detailed the definition of severe weather, including rainy, snowy and foggy days that are common in the port area;

[0079] Test items: timely identification of acceleration, deceleration, braking, etc. of the vehicle ahead and corresponding responses;

[0080] The test conditions specify the behavior of the target vehicle in front of the main vehicle, including the behavior of the target vehicle in the same lane or in different lanes that can affect the decision-making of the main vehicle, including: sudden acceleration of the front vehicle, sudden deceleration of the front vehicle, sudden braking of the front vehicle, lane change of the front vehicle, the front vehicle cutting in, a stationary vehicle starting and cutting in, vehicles on both sides of an idle lane merging into the lane at the same time, a stationary vehicle after the target vehicle cuts out, and an oncoming vehicle turning around.

[0081] Specific implementation example:

[0082] (I) Test preparation

[0083] A simulation environment based on a typical port layout is established, which includes multiple berths, container yards, warehouses, and the road network connecting them.

[0084] Import the model and hardware model of a port autonomous driving vehicle to be tested into the simulation environment.

[0085] Select the path planning accuracy test item and loading and unloading action accuracy test item in the functional test type.

[0086] (II) Test condition setting

[0087] Traffic flow is set up in the port simulation environment, including a small number of other auxiliary operation vehicles (such as forklifts) driving around the yard with a speed range of [30-40km / h].

[0088] Simulate sunny weather conditions with moderate light intensity and no interfering factors such as rain and fog.

[0089] (III) Test Execution

[0090] Start the simulation and watch the self-driving truck depart from the yard and follow the preset path to the berth for loading operations. Record the vehicle's position information during the driving process to check whether it deviates from the planned path.

[0091] At the berth, observe the interaction between the vehicle and the loading and unloading equipment, check whether the vehicle is accurately positioned for loading, and whether the loading action complies with standard operating procedures.

[0092] After loading is completed, the vehicle automatically drives to the designated warehouse to unload, and the accuracy of the driving route and unloading action is also recorded.

[0093] (IV) Results Analysis

[0094] The collected vehicle position data is processed to calculate the deviation between the actual vehicle path and the planned path. If the deviation is within the preset range (e.g., less than 10 cm), the path planning accuracy test item is preliminarily passed.

[0095] Check whether there are any operational errors or abnormalities based on the records of loading and unloading actions. If the loading and unloading actions are smooth and completed accurately, the loading and unloading action accuracy test item is passed.

[0096] Record the test results in the test report. If there are any problems, analyze the causes of the problems and make suggestions for improvement.

[0097] Through the above steps, the simulation test of the port autonomous driving vehicle under specific test types, items and working conditions can be completed. Repeating the above process can conduct a comprehensive evaluation of the vehicle for different test combinations.

[0098] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A simulation test method for an autonomous driving vehicle in a port, characterized by: The following steps are involved: Step 1: Analyze the test requirements. Analyze the test requirements according to the test procedures for the automatic driving function of intelligent connected vehicles, determine the type of test, the test items to be selected, and the test conditions; Step 2: Port and vehicle modeling: Establish a simulation model based on a typical port layout. The port model includes berths, container yards, warehouses, bridges, loading and unloading equipment, traffic signs, and the road network connecting them. Select the parameters of the test vehicle according to the needs, including vehicle size, vehicle weight, vehicle power, speed, acceleration, braking data, and power loss ratio, and import the port autonomous driving vehicle model to be tested; Step 3: Test preparation, loading simulation software modules, including scheduling task module, path planning module, data perception module, and decision-making module; Step 4: Test type selection. Test types include functional testing, performance testing, safety testing, reliability testing, and boundary testing. Select the test type according to the test requirements. Step 5: Determine the test items and select specific test items according to the test type; Step 6: Select the test condition. Select the test condition according to the test type and test items, including automatic lane change test condition, emergency braking test condition, and extreme weather test condition; Step 7: Execute simulation tests, and conduct tests in sequence according to the selected test type, test items, and test conditions, and collect the operating data of the unmanned vehicle during the test, including the vehicle's position, speed, acceleration, sensor data, and decision instructions; Step 8: Result analysis and evaluation: Analyze the collected data, compare the test indicators and acceptance criteria, determine the performance of the unmanned vehicle in each test, and generate a detailed test report to provide a basis for the actual application of autonomous vehicles in ports.

2. A port autonomous driving vehicle simulation test method according to claim 1, characterized in that: In step 5, each test type includes at least two test items, and each test item has a test indicator and an acceptance standard.

3. A port autonomous driving vehicle simulation test method according to claim 2, characterized in that: In step 5, 1 to 3 test items may be selected simultaneously for composite testing.

4. A port autonomous driving vehicle simulation test method according to claim 1, characterized in that: In step 6, 1 to 3 test conditions may be selected simultaneously for composite testing.

5. A port autonomous driving vehicle simulation test method according to claim 1, characterized in that: The content of the test report in step 8 includes a description of the test process, an analysis of the test results, existing problems and improvement suggestions.

Images