Method for testing and evaluating autonomous maneuvering function and performance of ground unmanned vehicle

Abstract

The invention belongs to the technical field of unmanned vehicle testing, and particularly relates to a ground unmanned vehicle autonomous maneuvering function and performance test and evaluation method, which comprises the following steps: step 1, setting a test scene; 2, collecting and processing test data; and 3, determining an evaluation index system. According to the invention, on the basis of unstructured environmental risk analysis, autonomous maneuvering function and performance testing and multi-dimensional index quantitative evaluation of the ground unmanned vehicle are realized, and effective reference can be provided for development and improvement of autonomous maneuvering characteristics of the tested vehicle.

Description

Technical Field

[0001] The present invention belongs to the technical field of unmanned vehicle testing, and in particular relates to a method for testing and evaluating the autonomous maneuvering function and performance of a ground unmanned vehicle. Background Art

[0002] In recent years, with the continuous development of artificial intelligence technology, unmanned vehicles have gradually become the focus of people's attention. Unlike civilian self-driving cars that mainly travel on structured roads in urban blocks, ground unmanned vehicles in the military field will not only face complex roads such as dirt roads, grasslands, gravel, deserts, jungles, etc. when autonomously maneuvering in the wild, but also encounter complex terrains such as longitudinal slopes, trenches, and bosses. In order to verify and improve the robustness and adaptability of the designed unmanned vehicle autonomous maneuvering algorithms, current research and testing work focuses on the performance improvement and verification of single algorithms of subsystems such as environmental perception algorithms, real-time positioning algorithms, behavioral decision algorithms, motion planning algorithms, and motion control algorithms. However, there is currently a lack of relevant theories and methods for systematic analysis, reasonable and reliable quantitative evaluation of the autonomous maneuvering functions and performance of the entire vehicle, and it is necessary to conduct research and design on this. Summary of the invention

[0003] 1. Technical issues to be resolved

[0004] The technical problem to be solved by the present invention is: how to provide a method for testing and evaluating the autonomous maneuvering function and performance of a ground unmanned vehicle.

[0005] (II) Technical solution

[0006] In order to solve the above technical problems, the present invention provides a method for testing and evaluating the autonomous maneuvering function and performance of a ground unmanned vehicle, the method comprising the following steps:

[0007] Step 1: Experimental scene setup;

[0008] Step 2: Experimental data collection and processing;

[0009] Step 3: Determine the evaluation index system.

[0010] (III) Beneficial effects

[0011] Compared with the prior art, the present invention aims at the needs of high-speed and highly adaptable autonomous maneuvering of ground unmanned vehicles in unstructured environments, and designs five test items from the two dimensions of maneuvering speed and passing ability; then designs the test conditions, test methods and test data processing strategies of each test item in detail; then, establishes a five-dimensional evaluation index system of autonomous maneuvering functions and performance based on the analysis of key functions and performance of autonomous maneuvering; finally, the index is dimensionlessly processed to achieve a comprehensive evaluation of the autonomous maneuvering functions and performance of ground unmanned vehicles. Based on the risk analysis of unstructured environments, the present invention realizes the autonomous maneuvering function and performance test of ground unmanned vehicles and the quantitative evaluation of multi-dimensional indicators, which can provide an effective reference for the development and improvement of the autonomous maneuvering performance of the tested vehicles. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1a The figure is a schematic diagram of a method for testing and evaluating the autonomous maneuvering function and performance of a ground unmanned vehicle suitable for unstructured environments.

[0013] Figure 1b This is the test equipment and installation drawing.

[0014] Figure 2 This is a simplified diagram of the autonomous average off-road speed test process for ground unmanned vehicles.

[0015] Figure 3 This is a simplified diagram of the ground unmanned vehicle climbing process.

[0016] Figure 4 This is a simplified diagram of the process of a ground unmanned vehicle autonomously crossing a trench.

[0017] Figure 5 This is a simplified diagram of the process of a ground unmanned vehicle autonomously passing through a boss.

[0018] Figure 6 This is a five-dimensional evaluation indicator system diagram.

[0019] Figure 7 It is a function and performance network diagram. DETAILED DESCRIPTION

[0020] In order to make the purpose, content, and advantages of the present invention more clear, the specific implementation methods of the present invention are further described in detail below in conjunction with the accompanying drawings and examples.

[0021] In order to solve the above technical problems, the present invention provides a method for testing and evaluating the autonomous maneuvering function and performance of a ground unmanned vehicle, such as Figure 1a As shown, the method comprises the following steps:

[0022] Step 1: Experimental scene setup;

[0023] Step 2: Experimental data collection and processing;

[0024] Step 3: Determine the evaluation index system.

[0025] Among them, Figure 1b As shown, step 1 includes test site setting, test task setting, test equipment installation and signal acquisition;

[0026] The test site setting includes: selecting an unstructured road including straights, curves, longitudinal ramps that vehicles can pass through at most, trenches that vehicles can pass through at most, and convex platforms that vehicles can pass through at most as the test site. The total length of the straights that vehicles can pass through should be no less than 3000m, and the straights should include paved roads, gravel roads and other complex road surfaces. The number of curves should be no less than 6 and at least one uphill, trench crossing and convex platform passing action should be completed. The road slope angle is determined by the maximum climbing grade of the test vehicle, the trench width is determined by the maximum trench crossing width of the test vehicle, and the convex platform height is the maximum convex platform height that the test vehicle can pass through.

[0027] The test tasks include: autonomous average off-road speed test, autonomous maximum off-road speed test, autonomous decision-making and control ability test for passing longitudinal slopes, autonomous decision-making and control ability test for passing trenches, and autonomous decision-making and control ability test for passing convex platforms;

[0028] The test equipment installation includes: installing a satellite signal receiver, an inertial navigation unit, and a VBOX tester on the vehicle under test;

[0029] Signal acquisition: collect the speed, acceleration, heading angle, pitch angle, yaw angle, angular velocity, longitude and latitude, and time information of the test vehicle;

[0030] Among them, acceleration collection involves acceleration in three directions: longitudinal, lateral and vertical.

[0031] Among them, the test data collection and processing of step 2 include: autonomous average off-road speed test of ground unmanned vehicles, autonomous maximum off-road speed test of ground unmanned vehicles, decision-making and control ability test for autonomous passing through longitudinal slopes, decision-making and control ability test for autonomous crossing trenches, and decision-making and control ability test for autonomous passing through bosses.

[0032] (I) Autonomous average off-road speed test of ground unmanned vehicles, such as Figure 2 Shown

[0033] (1) Test conditions:

[0034] 1) Test environment requirements: natural environmental conditions, wind speed no greater than 3.3m / s, including off-road distance of no less than 1 km;

[0035] 2) Technical requirements and quantity requirements for the test sample and accompanying test samples:

[0036] Tested object: Unmanned ground vehicle

[0037] Table 1 Technical requirements and quantity requirements for test products

[0038]

[0039] 3) Instruments and equipment:

[0040] High-precision VBOX: speed resolution 0.01km / h, distance resolution 1cm, time resolution 0.01s;

[0041] Spectrum analyzer: 3Hz~40GHz;

[0042] Anemometer: 0.1m / s;

[0043] (2) Test method:

[0044] 1) Use a short-range remote control to remotely control the ground unmanned vehicle to a designated location to collect and process the driving route, delete duplicate waypoints, and mark the starting point, intermediate point, and end point;

[0045] 2) Switch the authority to the control terminal; use the control terminal to connect to the unmanned vehicle, bind the driving route and send the autonomous maneuvering command, then use the test computer to remotely connect to the autonomous maneuvering unmanned vehicle, confirm the autonomous maneuvering route of the ground unmanned vehicle, and confirm that the chassis is in the autonomous maneuvering state;

[0046] 3) Release the stop button, the ground unmanned vehicle will autonomously maneuver along the designated path and maintain the autonomous driving state for 30 minutes before stopping;

[0047] (3) Data processing:

[0048] The total mileage is s, the starting time is t0, and the ending time is t1. The autonomous average off-road speed v is calculated according to the following formula;

[0049]

[0050] (II) Autonomous maximum off-road speed test of ground unmanned vehicles

[0051] (1) Test conditions:

[0052] 1) Test environment requirements: natural environmental conditions, wind speed no greater than 3.3m / s, road width no less than 5 meters

[0053] 2) Technical requirements and quantity requirements for the test sample and accompanying test samples:

[0054] Tested object: Unmanned ground vehicle

[0055] Table 2 Technical requirements and quantity requirements for test products

[0056]

[0057] 3) Instruments and equipment:

[0058] High-precision VBOX: speed resolution 0.01km / h, distance resolution 1cm, time resolution 0.01s;

[0059] Spectrum analyzer: 3Hz~40GHz;

[0060] Anemometer: 0.1m / s;

[0061] (2) Test method:

[0062] 1) Use a short-range remote control to remotely control the ground unmanned vehicle to a designated location to collect and process the driving route, delete duplicate waypoints, and mark the starting point, intermediate point, and end point;

[0063] 2) Switch the authority to the control terminal; use the control terminal to connect to the unmanned vehicle, bind the driving route and send the autonomous maneuvering command, then use the test computer to remotely connect to the autonomous maneuvering unmanned vehicle, confirm the autonomous maneuvering route of the ground unmanned vehicle, and confirm that the chassis is in the autonomous maneuvering state;

[0064] 3) Release the stop button, the ground unmanned vehicle will autonomously maneuver along the designated path, gradually accelerate to the maximum speed, maintain the maximum speed for 20 seconds, and then gradually decelerate to stop;

[0065] (3) Data processing:

[0066] 10 seconds of data is intercepted in the highest speed area of ​​the speed curve, and the lowest driving speed in 10 seconds is taken as the maximum autonomous speed of the ground unmanned vehicle.

[0067] (III) Decision-making and control ability test for autonomously passing the longitudinal slope, such as Figure 3 Shown

[0068] (1) Test conditions:

[0069] 1) Test environment requirements:

[0070] In natural environment, the wind speed is no more than 3.3m / s;

[0071] 2) Site conditions:

[0072] A longitudinal slope lower than the maximum design requirement;

[0073] 3) Technical requirements and quantity requirements for the test sample and accompanying test samples:

[0074] Tested object: Unmanned ground vehicle

[0075] Table 3 Technical requirements and quantity requirements for test products

[0076]

[0077]

[0078] (2) Test method:

[0079] 1) Use a short-range remote control to remotely control the ground unmanned vehicle to the designated site and align it with the ramp, 2 meters in front of the bottom of the ramp;

[0080] 2) Switch the authority to the control terminal; use the control terminal to connect to the vehicle, send the autonomous obstacle crossing command, and then use the test computer to remotely connect to the autonomous mobile unmanned vehicle to confirm that the ground unmanned vehicle is in the autonomous obstacle crossing state;

[0081] 3) Release the stop button, and the ground unmanned vehicle will perform autonomous maneuvers. During the autonomous maneuvers, the test computer will be used to observe the vehicle speed, curvature, and obstacle crossing status in real time, and record the results;

[0082] (3) Data processing:

[0083] Record the speed, curvature, oil leakage and load fixation of the vehicle when it autonomously passes through the longitudinal slope.

[0084] (IV) Autonomous trench crossing decision-making and control capability test, such as Figure 4 Shown

[0085] (1) Test conditions:

[0086] 1) Test environment: natural environmental conditions;

[0087] 2) Site conditions:

[0088] (a) A trench of maximum design width, not less than 2 times the vehicle length fore and aft and 2 times the vehicle width for lateral;

[0089] (b) A trench exceeding the maximum design width, which is not less than 2 times the vehicle length fore and aft and 2 times the vehicle width for left and right;

[0090] 3) Technical requirements and quantity requirements for the test sample and accompanying test samples:

[0091] Tested object: Unmanned ground vehicle

[0092] Table 4 Technical requirements and quantity requirements for test products

[0093]

[0094] (2) Test method:

[0095] 1) Use a short-range remote control to remotely control the ground unmanned vehicle to the designated site and align it with the trench, 2 meters away from the obstacle;

[0096] 2) Switch the authority to the control terminal; use the control terminal to connect to the vehicle, send the autonomous obstacle crossing command, and then use the test computer to remotely connect to the autonomous mobile unmanned vehicle to confirm that the ground unmanned vehicle is in the autonomous obstacle crossing state;

[0097] 3) Release the stop button, and the ground unmanned vehicle will autonomously cross obstacles. During the autonomous obstacle crossing process, the test computer is used to observe the vehicle speed, curvature and obstacle crossing status in real time, and record the results;

[0098] (3) Data processing:

[0099] The process data of autonomous trench crossing are recorded by the control end and the unmanned vehicle respectively; the process data include expected speed, actual speed, expected curvature, actual curvature, acceleration of each axis and angular acceleration.

[0100] (V) Autonomously pass the decision-making and control capability test of the boss, such as Figure 5 Shown

[0101] (1) Test conditions:

[0102] 1) Test environment: natural environmental conditions;

[0103] 2) Site conditions:

[0104] (a) One boss of maximum design height, with a boss length not less than 2 times the vehicle length;

[0105] (b) A boss that exceeds the maximum design height and has a length not less than twice the vehicle length;

[0106] 3) Technical requirements and quantity requirements for the test sample and accompanying test samples:

[0107] Tested object: Unmanned ground vehicle

[0108] Table 5 Technical requirements and quantity requirements for test products

[0109]

[0110] (2) Test method:

[0111] 1) Use a short-range remote control to remotely control the heavy ground unmanned vehicle to the designated site and align it with the convex platform, 2 meters away from the obstacle;

[0112] 2) Switch the authority to the control terminal; use the control terminal to connect to the vehicle, send the autonomous obstacle crossing command, and then use the test computer to remotely connect to the autonomous mobile unmanned vehicle to confirm that the ground unmanned vehicle is in the autonomous obstacle crossing state;

[0113] 3) Release the stop button, and the ground unmanned vehicle will autonomously cross obstacles. During the autonomous obstacle crossing process, the test computer is used to observe the vehicle speed, curvature and obstacle crossing status in real time, and record the results;

[0114] (3) Data processing:

[0115] The process data of autonomous crossing over the boss and trench are recorded by the control end and the unmanned vehicle respectively; the process data include expected speed, actual speed, expected curvature, actual curvature, acceleration of each axis and angular acceleration.

[0116] Among them, in the step 3, according to the requirements of the functions and performance indicators of the autonomous maneuvering of the unmanned vehicle, an evaluation index system can be constructed, including: the average speed of autonomous maneuvering, the maximum speed of autonomous maneuvering, the ability to climb the slope autonomously, the ability to cross the trench autonomously, and the ability to pass the convex platform autonomously, such as Figure 6 As shown;

[0117] (1) Average speed of autonomous maneuvers

[0118] Obtained by following the total mileage and total time conversion, as follows:

[0119]

[0120] (2) Maximum speed of autonomous maneuvering

[0121] Take 10 seconds of data in the highest speed area of ​​the speed curve and take the lowest driving speed in 10 seconds, which is the maximum autonomous speed v of the ground unmanned vehicle. max ;

[0122] (3) Autonomous climbing capability

[0123] The autonomous climbing ability evaluation index is designed based on the following three aspects:

[0124] 1) Repeat the test for a set number of times, and each time the longitudinal slope should be determined to be a passable longitudinal slope; the number of times it is determined to be a passable longitudinal slope is Count_ACLCP;

[0125] 2) When the correct decision is to pass, the vehicle can safely pass the longitudinal slope; the number of times the vehicle safely passes the longitudinal slope is recorded as Count_BCLCP;

[0126] 3) When the vehicle slips during safe passage through the longitudinal slope, the slip distance shall not exceed the designed distance; the number of times the slip distance is not exceeded is recorded as Count_CCLCP;

[0127] The evaluation index of autonomous climbing ability is:

[0128] CLCP=Count_ACLCP+Count_BCLCP+Count_CCLCP

[0129] (4) Autonomous trench crossing capability

[0130] The evaluation index of autonomous trench crossing capability is designed based on the following two aspects:

[0131] 1) Repeat the test for a set number of times, and each time it should be possible to correctly decide which of the two trenches can be passed; the number of correct decisions on whether it can be passed is Count_ATCCP;

[0132] 2) When the correct decision is to pass, the player can safely pass through the trench. The number of times the player safely crosses the trench is recorded as Count_BTCCP;

[0133] The evaluation indicators of autonomous trench crossing capability are:

[0134] TCCP=Count_ATCCP+Count_BTCCP

[0135] (5) Ability to pass through bosses autonomously

[0136] The evaluation index of the ability to design self-passing bosses is considered from the following two aspects:

[0137] 1) Repeat the test for a set number of times, and each time it should be possible to correctly decide which of the two bosses can pass; the number of correct decisions on whether it can pass is Count_ACCPCP;

[0138] 2) When the correct decision is to pass, the vehicle can pass the boss safely; the number of times the vehicle passes the boss safely is recorded as Count_BCCPCP;

[0139] The evaluation index of the ability to pass bosses autonomously is:

[0140] CCPCP=Count_ACCPCP+Count_BCCPCP.

[0141] (6) Comprehensive evaluation indicators

[0142] Each index is dimensionless, that is, the autonomous average speed is divided by the designed autonomous average speed, the autonomous maximum speed is divided by the designed autonomous maximum speed, the autonomous climbing ability, the autonomous trench crossing ability, and the autonomous convex platform passing ability are divided by the corresponding test times, thus obtaining the following: Figure 7 The function and performance network diagram shown is used to evaluate the autonomous maneuvering functions and performance of ground unmanned vehicles.

[0143] Example 1

[0144] Aiming at the demand for high-speed and highly adaptable autonomous maneuvering of ground unmanned vehicles in unstructured environments, the present invention designs five test items from two dimensions of maneuvering speed and passing ability; then designs the test conditions, test methods and test data processing strategies for each test item in detail; then, based on the analysis of key functions and performances of autonomous maneuvering, establishes a five-dimensional evaluation index system for autonomous maneuvering functions and performances; finally, the indicators are dimensionlessly processed to achieve a comprehensive evaluation of the autonomous maneuvering functions and performances of ground unmanned vehicles.

[0145] First, a satellite signal receiver, an inertial navigation device and a VBOX tester are installed on the vehicle under test to mainly collect information such as the speed, acceleration (longitudinal, lateral, vertical), heading angle, pitch angle, yaw angle, angular velocity, longitude and latitude, and time of the vehicle under test.

[0146] Then, an unstructured road including straight roads, curves, longitudinal ramps that vehicles can pass through, ditches that vehicles can pass through and those that they cannot pass through, and bosses that vehicles can pass through and those that they cannot pass through were selected as the test site. The maximum speed, average speed, autonomous climbing of longitudinal slopes, autonomous crossing of ditches, and autonomous passing of bosses were completed respectively, and the test data on the road was collected.

[0147] On this basis, according to the five-dimensional evaluation index system, the test data is calculated to obtain the autonomous average speed, autonomous maximum speed, autonomous climbing ability, autonomous trench crossing ability, and autonomous boss passing ability functions and performance indicators.

[0148] Finally, the five evaluation indicators were dimensionlessly processed to obtain the comprehensive evaluation results of autonomous maneuvering functions and performances.

[0149] Example 2

[0150] This embodiment includes the following contents:

[0151] Step 1: Test equipment installation

[0152] A satellite signal receiver, an inertial navigation device and a VBOX tester are installed on the ground unmanned vehicle under test.

[0153] Step 2: Select a test site that meets the test requirements and prepare the test products, test samples and equipment

[0154] 2.1 Autonomous average off-road speed test of ground unmanned vehicles

[0155] Test conditions

[0156] (a) Test environment requirements: natural environment conditions, wind speed no greater than 3.3m / s, including off-road area of ​​no less than 1 km

[0157] (b) Technical requirements and quantity requirements for the test sample and accompanying test samples:

[0158] Tested object: Unmanned ground vehicle

[0159] Table 1 Technical requirements and quantity requirements for test products

[0160]

[0161] (c) Instruments and equipment: high-precision VBOX (velocity resolution 0.01 km / h, distance resolution 1 cm, time resolution 0.01 s), spectrum analyzer (3 Hz ~ 40 GHz), anemometer (0.1 m / s).

[0162] 2.2 Autonomous maximum off-road speed test of ground unmanned vehicles

[0163] Test conditions

[0164] (a) Test environment requirements: natural environmental conditions, wind speed no greater than 3.3m / s, road width no less than 5 meters

[0165] (b) Technical requirements and quantity requirements for the test sample and accompanying test samples:

[0166] Tested object: Unmanned ground vehicle

[0167] Table 2 Technical requirements and quantity requirements for test products

[0168]

[0169]

[0170] (c) Instruments and equipment: high-precision VBOX (velocity resolution 0.01 km / h, distance resolution 1 cm, time resolution 0.01 s), spectrum analyzer (3 Hz ~ 40 GHz), anemometer (0.1 m / s).

[0171] 2.3 Decision-making and control ability test for autonomously passing the longitudinal slope

[0172] Test conditions

[0173] (a) Test environment requirements:

[0174] In natural environment, the wind speed shall not exceed 3.3m / s.

[0175] (b) Site conditions:

[0176] Longitudinal slope lower than the maximum design requirement.

[0177] (c) Technical requirements and quantity requirements for the test sample and accompanying test samples:

[0178] Tested object: Unmanned ground vehicle

[0179] Table 3 Technical requirements and quantity requirements for test products

[0180]

[0181]

[0182] 2.4 Decision-making and control capability test of autonomous trench crossing

[0183] Test conditions

[0184] (a) Test environment:

[0185] Natural environmental conditions.

[0186] (b) Site conditions:

[0187] (1) A trench of maximum design width, not less than 2 times the vehicle length in front and rear and 2 times the vehicle width in left and right.

[0188] (2) A trench that exceeds the maximum design width. The trench should be no less than 2 times the length of the vehicle in front and back and no less than 2 times the width of the vehicle in the left and right directions.

[0189] (c) Technical requirements and quantity requirements for the test sample and accompanying test samples:

[0190] Tested object: Unmanned ground vehicle

[0191] Table 4 Technical requirements and quantity requirements for test products

[0192]

[0193]

[0194] 2.5 Autonomously pass the decision-making and control ability test of the boss

[0195] Test conditions

[0196] (a) Test environment:

[0197] Natural environmental conditions.

[0198] (b) Site conditions:

[0199] (1) One boss with maximum design height, the boss length shall not be less than 2 times the vehicle length.

[0200] (2) One boss that exceeds the maximum design height, and the boss length is not less than 2 times the vehicle length. (c) Technical requirements and quantity requirements for the test product and accompanying test products:

[0201] Tested object: Unmanned ground vehicle

[0202] Table 5 Technical requirements and quantity requirements for test products

[0203]

[0204]

[0205] Step 3: Carry out each test item according to the test method and collect and process data

[0206] 3.1 Autonomous average off-road speed test of ground unmanned vehicles

[0207] Test Method

[0208] 1. Use a short-range remote control to remotely control the ground unmanned vehicle to a designated location to collect and process the driving route, delete duplicate waypoints, and mark the starting point, intermediate point, and end point;

[0209] 2. Switch permissions to the control terminal. Use the control terminal to connect to the unmanned vehicle, bind the driving route and send the autonomous maneuvering command, then use the test computer to remotely connect to the autonomous maneuvering unmanned vehicle, confirm the autonomous maneuvering route of the ground unmanned vehicle, and confirm that the chassis is in the autonomous maneuvering state;

[0210] 3. Release the stop button, and the ground unmanned vehicle will autonomously maneuver along the designated path and maintain the autonomous driving state for 30 minutes before stopping.

[0211] Data processing

[0212] The total mileage is s, the starting time is t0, and the ending time is t1. The autonomous average off-road speed v is calculated according to the following formula.

[0213]

[0214] 3.2 Autonomous maximum off-road speed test of ground unmanned vehicles

[0215] Test Method

[0216] 1. Use a short-range remote control to remotely control the ground unmanned vehicle to a designated location to collect and process the driving route, delete duplicate waypoints, and mark the starting point, intermediate point, and end point;

[0217] 2. Switch permissions to the control terminal. Use the control terminal to connect to the unmanned vehicle, bind the driving route and send the autonomous maneuvering command, then use the test computer to remotely connect to the autonomous maneuvering unmanned vehicle, confirm the autonomous maneuvering route of the ground unmanned vehicle, and confirm that the chassis is in the autonomous maneuvering state;

[0218] 3. Release the stop button, and the ground unmanned vehicle will autonomously maneuver along the designated path, gradually accelerate to the maximum speed, maintain the maximum speed for 20 seconds, and then gradually decelerate to a stop.

[0219] Data processing

[0220] 10 seconds of data is intercepted in the highest speed area of ​​the speed curve, and the lowest driving speed in 10 seconds is taken as the maximum autonomous speed of the ground unmanned vehicle.

[0221] 3.3 Decision-making and control ability test for autonomously passing the longitudinal slope

[0222] Test Method

[0223] 1. Use a short-range remote control to remotely control the ground unmanned vehicle to the designated site and align it with the ramp, 2 meters in front of the bottom of the ramp;

[0224] 2. Switch permissions to the control terminal. Use the control terminal to connect to the vehicle, send the autonomous obstacle crossing command, and then use the test computer to remotely connect to the autonomous mobile unmanned vehicle to confirm that the ground unmanned vehicle is in the autonomous obstacle crossing state;

[0225] 3. Release the stop button and the ground unmanned vehicle will perform autonomous maneuvers. During the autonomous maneuvers, the test computer will be used to observe the vehicle speed, curvature and obstacle crossing status in real time and record the results.

[0226] Data processing

[0227] Record the vehicle's autonomous passage through the longitudinal slope (speed, curvature), oil leakage, and load fixation.

[0228] 3.4 Decision-making and control capability test of autonomous trench crossing

[0229] Test Method

[0230] 1. Use a short-range remote control to remotely control the ground unmanned vehicle to the designated site and align it with the trench, 2 meters away from the obstacle;

[0231] 2. Switch permissions to the control terminal. Use the control terminal to connect to the vehicle, send the autonomous obstacle crossing command, and then use the test computer to remotely connect to the autonomous mobile unmanned vehicle to confirm that the ground unmanned vehicle is in the autonomous obstacle crossing state;

[0232] 3. Release the stop button, and the ground unmanned vehicle will autonomously cross obstacles. During the autonomous obstacle crossing process, use the test computer to observe the vehicle speed, curvature and obstacle crossing status in real time, and record the results.

[0233] Data processing

[0234] The process data of autonomous trench crossing (expected speed, actual speed, expected curvature, actual curvature, acceleration of each axis and angular acceleration) are recorded by the control end and the unmanned vehicle respectively.

[0235] 3.5 Autonomous decision-making and control capability test of the boss

[0236] Test Method

[0237] 1. Use a short-range remote control to remotely control the heavy ground unmanned vehicle to the designated site and align it with the convex platform, 2 meters away from the obstacle;

[0238] 2. Switch permissions to the control terminal. Use the control terminal to connect to the vehicle, send the autonomous obstacle crossing command, and then use the test computer to remotely connect to the autonomous mobile unmanned vehicle to confirm that the ground unmanned vehicle is in the autonomous obstacle crossing state;

[0239] 3. Release the stop button, and the ground unmanned vehicle will autonomously cross obstacles. During the autonomous obstacle crossing process, use the test computer to observe the vehicle speed, curvature and obstacle crossing status in real time, and record the results.

[0240] Data processing

[0241] The process data of autonomous crossing over the boss and trench (expected speed, actual speed, expected curvature, actual curvature, acceleration of each axis and angular acceleration) are recorded by the control end and the unmanned vehicle respectively.

[0242] Step 4: Calculation of five-dimensional evaluation indicators

[0243] 4.1 Average speed of autonomous maneuvers

[0244] It can be obtained by converting the total mileage and total time as follows:

[0245]

[0246] 4.2 Maximum speed of autonomous maneuvering

[0247] Take 10 seconds of data in the highest speed area of ​​the speed curve and take the lowest driving speed in 10 seconds, which is the maximum autonomous speed v of the ground unmanned vehicle. max .

[0248] 4.3 Autonomous climbing capability

[0249] The autonomous climbing ability evaluation index is designed based on the following three aspects.

[0250] 1. Repeat the test for a set number of times, and each time the slope should be determined to be passable. The number of times it is determined to be passable is Count_ACLCP.

[0251] 2. When the correct decision is to pass, the vehicle can safely pass the longitudinal slope. The number of times the vehicle safely passes the longitudinal slope is recorded as Count_BCLCP.

[0252] 3. When the vehicle slips during safe passage through the longitudinal slope, the slip distance does not exceed the designed distance. The number of times the slip distance is not exceeded is recorded as Count_CCLCP.

[0253] The evaluation index of autonomous climbing ability is:

[0254] CLCP=Count_ACLCP+Count_BCLCP+Count_CCLCP

[0255] 4.4 Autonomous Trench Crossing Capability

[0256] The evaluation index of autonomous trench crossing capability is designed based on the following two aspects.

[0257] 1. Repeat the test for a set number of times, and each time you should be able to correctly decide which of the two trenches you can pass through. The number of correct decisions on whether you can pass through is Count_ATCCP.

[0258] 2. When the correct decision is to pass, the player can safely pass through the trench. The number of times the player safely crosses the trench is recorded as Count_BTCCP.

[0259] The evaluation indicators of autonomous trench crossing capability are:

[0260] TCCP=Count_ATCCP+Count_BTCCP

[0261] 4.5 Autonomous passing through boss capability

[0262] The design independently passes the boss capability evaluation index from the following two aspects.

[0263] 1. Repeat the test for a set number of times, and each time you should be able to correctly decide which of the two bosses can pass. The number of correct decisions on whether it can pass is Count_ACCPCP.

[0264] 2. When the correct decision is to pass, the vehicle can pass the boss safely. The number of times the vehicle passes the boss safely is recorded as Count_BCCPCP.

[0265] The evaluation index of the ability to pass bosses autonomously is:

[0266] CCPCP=Count_ACCCPP+Count_BCCPCP

[0267] Step 5: Comprehensive evaluation indicators

[0268] After dimensionless processing of each index, that is, autonomous average speed divided by designed autonomous average speed, autonomous maximum speed divided by designed autonomous maximum speed, autonomous climbing ability, autonomous trench crossing ability, and autonomous convex platform passing ability divided by the corresponding test times, we can get the following: Figure 7 The function and performance network diagram shown is used to evaluate the autonomous maneuvering functions and performance of ground unmanned vehicles.

[0269] In summary, the present invention belongs to the field of unmanned vehicle testing technology, and specifically relates to a method for testing and evaluating the autonomous maneuvering function and performance of a ground unmanned vehicle. First, a satellite signal receiver, an inertial navigation device and a VBOX tester are installed on the vehicle under test, and the speed, acceleration (longitudinal, lateral, vertical), heading angle, pitch angle, yaw angle, angular velocity, longitude and latitude, and time of the vehicle under test are mainly collected. Then, an unstructured road including a straight road, a curve, a longitudinal ramp that the vehicle can pass through, a trench that the vehicle can pass through and cannot pass through, and a convex platform that the vehicle can pass through and cannot pass through are selected as the test site, and the maximum speed, average speed, autonomous longitudinal slope climbing, autonomous crossing of trenches, and autonomous passing of convex platforms are completed respectively, and the test data on the road are collected. On this basis, a five-dimensional evaluation index system based on key functions and performance indicators is proposed, including autonomous maximum speed, autonomous average speed, autonomous longitudinal slope climbing, autonomous crossing of trenches, and autonomous passing of convex platforms. Finally, the five evaluation indicators are dimensionlessly processed to obtain the comprehensive evaluation results of autonomous maneuvering performance. Based on unstructured environmental risk analysis, the present invention realizes the autonomous maneuvering function and performance testing of ground unmanned vehicles and quantitative evaluation of multi-dimensional indicators, which can provide an effective reference for the development and improvement of the autonomous maneuvering performance of the tested vehicles.

[0270] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the technical principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A method for testing and evaluating the autonomous maneuvering function and performance of a ground unmanned vehicle, characterized in that: The method comprises the following steps: Step 1: Experimental scene setup; Step 2: Experimental data collection and processing; Step 3: Determine the evaluation index system.

2. The method for testing and evaluating the autonomous maneuvering function and performance of a ground unmanned vehicle according to claim 1, characterized in that: The step 1 includes test site setting, test task setting, test equipment installation and signal acquisition; The test site setting includes: selecting an unstructured road including straights, curves, longitudinal ramps that vehicles can pass through at most, trenches that vehicles can pass through at most, and convex platforms that vehicles can pass through at most as the test site. The total length of the straights that vehicles can pass through should be no less than 3000m, and the straights should include paved roads, gravel roads and other complex road surfaces. The number of curves should be no less than 6 and at least one uphill, trench crossing and convex platform passing action should be completed. The road slope angle is determined by the maximum climbing grade of the test vehicle, the trench width is determined by the maximum trench crossing width of the test vehicle, and the convex platform height is the maximum convex platform height that the test vehicle can pass through. The test tasks include: autonomous average off-road speed test, autonomous maximum off-road speed test, autonomous decision-making and control ability test for passing longitudinal slopes, autonomous decision-making and control ability test for passing trenches, and autonomous decision-making and control ability test for passing convex platforms; The test equipment installation includes: installing a satellite signal receiver, an inertial navigation unit, and a VBOX tester on the vehicle under test; Signal acquisition: collect the speed, acceleration, heading angle, pitch angle, yaw angle, angular velocity, longitude and latitude, and time information of the test vehicle; Among them, acceleration collection involves acceleration in three directions: longitudinal, lateral and vertical.

3. The method for testing and evaluating the autonomous maneuvering function and performance of a ground unmanned vehicle as claimed in claim 2, characterized in that: The test data collection and processing of step 2 include: autonomous average off-road speed test of ground unmanned vehicles, autonomous maximum off-road speed test of ground unmanned vehicles, autonomous decision-making and control ability test for passing longitudinal slopes, autonomous decision-making and control ability test for crossing trenches, and autonomous decision-making and control ability test for passing bosses.

4. The method for testing and evaluating the autonomous maneuvering function and performance of a ground unmanned vehicle as claimed in claim 3, characterized in that: In the step 2, (I) Autonomous average off-road speed test of ground unmanned vehicles (1) Test conditions: 1) Test environment requirements: natural environmental conditions, wind speed no greater than 3.3m / s, including off-road distance of no less than 1 km; 2) Technical requirements and quantity requirements for the test sample and accompanying test samples: Tested object: Unmanned ground vehicle Table 1 Technical requirements and quantity requirements for test products 3) Instruments and equipment: High-precision VBOX: speed resolution 0.01km / h, distance resolution 1cm, time resolution 0.01s; Spectrum analyzer: 3Hz~40GHz; Anemometer: 0.1m / s; (2) Test method: 1) Use a short-range remote control to remotely control the ground unmanned vehicle to a designated location to collect and process the driving route, delete duplicate waypoints, and mark the starting point, intermediate point, and end point; 2) Switch the authority to the control terminal; use the control terminal to connect to the unmanned vehicle, bind the driving route and send the autonomous maneuvering command, then use the test computer to remotely connect to the autonomous maneuvering unmanned vehicle, confirm the autonomous maneuvering route of the ground unmanned vehicle, and confirm that the chassis is in the autonomous maneuvering state; 3) Release the stop button, the ground unmanned vehicle will autonomously maneuver along the designated path and maintain the autonomous driving state for 30 minutes before stopping; (3) Data processing: The total mileage is s, the starting time is t0, and the ending time is t1. The autonomous average off-road speed v is calculated according to the following formula; 5. The method for testing and evaluating the autonomous maneuvering function and performance of a ground unmanned vehicle according to claim 4, characterized in that: In the step 2, (II) Autonomous maximum off-road speed test of ground unmanned vehicles (1) Test conditions: 1) Test environment requirements: natural environmental conditions, wind speed no greater than 3.3m / s, road width no less than 5 meters 2) Technical requirements and quantity requirements for the test sample and accompanying test samples: Tested object: Unmanned ground vehicle Table 2 Technical requirements and quantity requirements for test products 3) Instruments and equipment: High-precision VBOX: speed resolution 0.01km / h, distance resolution 1cm, time resolution 0.01s; Spectrum analyzer: 3Hz~40GHz; Anemometer: 0.1m / s; (2) Test method: 1) Use a short-range remote control to remotely control the ground unmanned vehicle to a designated location to collect and process the driving route, delete duplicate waypoints, and mark the starting point, intermediate point, and end point; 2) Switch the authority to the control terminal; use the control terminal to connect to the unmanned vehicle, bind the driving route and send the autonomous maneuvering command, then use the test computer to remotely connect to the autonomous maneuvering unmanned vehicle, confirm the autonomous maneuvering route of the ground unmanned vehicle, and confirm that the chassis is in the autonomous maneuvering state; 3) Release the stop button, the ground unmanned vehicle will autonomously maneuver along the designated path, gradually accelerate to the maximum speed, maintain the maximum speed for 20 seconds, and then gradually decelerate to stop; (3) Data processing: 10 seconds of data is intercepted in the highest speed area of ​​the speed curve, and the lowest driving speed in 10 seconds is taken as the maximum autonomous speed of the ground unmanned vehicle.

6. The method for testing and evaluating the autonomous maneuvering function and performance of a ground unmanned vehicle as claimed in claim 5, characterized in that: In the step 2, (III) Decision-making and control ability test for autonomously passing longitudinal slopes (1) Test conditions: 1) Test environment requirements: In natural environment, the wind speed is no more than 3.3m / s; 2) Site conditions: A longitudinal slope lower than the maximum design requirement; 3) Technical requirements and quantity requirements for the test sample and accompanying test samples: Tested object: Unmanned ground vehicle Table 3 Technical requirements and quantity requirements for test products (2) Test method: 1) Use a short-range remote control to remotely control the ground unmanned vehicle to the designated site and align it with the ramp, 2 meters in front of the bottom of the ramp; 2) Switch the authority to the control terminal; use the control terminal to connect to the vehicle, send the autonomous obstacle crossing command, and then use the test computer to remotely connect to the autonomous mobile unmanned vehicle to confirm that the ground unmanned vehicle is in the autonomous obstacle crossing state; 3) Release the stop button, and the ground unmanned vehicle will perform autonomous maneuvers. During the autonomous maneuvers, the test computer will be used to observe the vehicle speed, curvature, and obstacle crossing status in real time, and record the results; (3) Data processing: Record the speed, curvature, oil leakage and load fixation of the vehicle when it autonomously passes through the longitudinal slope.

7. The method for testing and evaluating the autonomous maneuvering function and performance of a ground unmanned vehicle according to claim 6, characterized in that: In the step 2, (IV) Decision-making and control capability test of autonomous trench crossing (1) Test conditions: 1) Test environment: natural environmental conditions; 2) Site conditions: (a) A trench of maximum design width, not less than 2 times the vehicle length fore and aft and 2 times the vehicle width for lateral; (b) A trench exceeding the maximum design width, which is not less than 2 times the vehicle length fore and aft and 2 times the vehicle width for left and right; 3) Technical requirements and quantity requirements for the test sample and accompanying test samples: Tested object: Unmanned ground vehicle Table 4 Technical requirements and quantity requirements for test products (2) Test method: 1) Use a short-range remote control to remotely control the ground unmanned vehicle to the designated site and align it with the trench, 2 meters away from the obstacle; 2) Switch the authority to the control terminal; use the control terminal to connect to the vehicle, send the autonomous obstacle crossing command, and then use the test computer to remotely connect to the autonomous mobile unmanned vehicle to confirm that the ground unmanned vehicle is in the autonomous obstacle crossing state; 3) Release the stop button, and the ground unmanned vehicle will autonomously cross obstacles. During the autonomous obstacle crossing process, the test computer is used to observe the vehicle speed, curvature and obstacle crossing status in real time, and record the results; (3) Data processing: The process data of autonomous trench crossing are recorded by the control end and the unmanned vehicle respectively; the process data include expected speed, actual speed, expected curvature, actual curvature, acceleration of each axis and angular acceleration.

8. The method for testing and evaluating the autonomous maneuvering function and performance of a ground unmanned vehicle according to claim 7, characterized in that: In the step 2, (V) Autonomous decision-making and control capability test of the boss (1) Test conditions: 1) Test environment: natural environmental conditions; 2) Site conditions: (a) One boss of maximum design height, with a boss length not less than 2 times the vehicle length; (b) A boss that exceeds the maximum design height and has a length not less than twice the vehicle length; 3) Technical requirements and quantity requirements for the test sample and accompanying test samples: Tested object: Unmanned ground vehicle Table 5 Technical requirements and quantity requirements for test products (2) Test method: 1) Use a short-range remote control to remotely control the heavy ground unmanned vehicle to the designated site and align it with the convex platform, 2 meters away from the obstacle; 2) Switch the authority to the control terminal; use the control terminal to connect to the vehicle, send the autonomous obstacle crossing command, and then use the test computer to remotely connect to the autonomous mobile unmanned vehicle to confirm that the ground unmanned vehicle is in the autonomous obstacle crossing state; 3) Release the stop button, and the ground unmanned vehicle will autonomously cross obstacles. During the autonomous obstacle crossing process, the test computer is used to observe the vehicle speed, curvature and obstacle crossing status in real time, and record the results; (3) Data processing: The process data of autonomous crossing over the boss and trench are recorded by the control end and the unmanned vehicle respectively; the process data include expected speed, actual speed, expected curvature, actual curvature, acceleration of each axis and angular acceleration.

9. The method for testing and evaluating the autonomous maneuvering function and performance of a ground unmanned vehicle as claimed in claim 8, characterized in that: In step 3, according to the requirements of the functions and performance indicators of the autonomous maneuvering of the unmanned vehicle, an evaluation index system can be constructed, including: the average speed of autonomous maneuvering, the maximum speed of autonomous maneuvering, the ability to climb the longitudinal slope autonomously, the ability to cross the trench autonomously, and the ability to pass the boss autonomously; (1) Average speed of autonomous maneuvers Obtained by following the total mileage and total time conversion, as follows: (2) Maximum speed of autonomous maneuvering Take 10 seconds of data in the highest speed area of ​​the speed curve and take the lowest driving speed in 10 seconds, which is the maximum autonomous speed v of the ground unmanned vehicle. max ; (3) Autonomous climbing capability The autonomous climbing ability evaluation index is designed based on the following three aspects: 1) Repeat the test for a set number of times, and each time the longitudinal slope should be determined to be a passable longitudinal slope; the number of times it is determined to be a passable longitudinal slope is Count_ACLCP; 2) When the correct decision is to pass, the vehicle can safely pass the longitudinal slope; the number of times the vehicle safely passes the longitudinal slope is recorded as Count_BCLCP; 3) When the vehicle slips during safe passage through the longitudinal slope, the slip distance shall not exceed the designed distance; the number of times the slip distance is not exceeded is recorded as Count_CCLCP; The evaluation index of autonomous climbing ability is: CLCP=Count_ACLCP+Count_BCLCP+Count_CCLCP (4) Autonomous trench crossing capability The evaluation index of autonomous trench crossing capability is designed based on the following two aspects: 1) Repeat the test for a set number of times, and each time it should be able to correctly decide which of the two trenches can be passed; the number of correct decisions on whether it can be passed is Count_ATCCP; 2) When the correct decision is to pass, the player can safely pass through the trench. The number of times the player safely crosses the trench is recorded as Count_BTCCP; The evaluation indicators of autonomous trench crossing capability are: TCCP=Count_ATCCP+Count_BTCCP (5) Ability to pass through bosses autonomously The evaluation index of the ability to design self-passing bosses is considered from the following two aspects: 1) Repeat the test for a set number of times, and each time it should be possible to correctly decide which of the two bosses can pass; the number of correct decisions on whether it can pass is Count_ACCPCP; 2) When the correct decision is to pass, the vehicle can pass the boss safely; the number of times the vehicle passes the boss safely is recorded as Count_BCCPCP; The evaluation index of the ability to pass bosses autonomously is: CCPCP=Count_ACCPCP+Count_BCCPCP.

10. The method for testing and evaluating the autonomous maneuvering function and performance of a ground unmanned vehicle according to claim 9, characterized in that: The step 3 further includes: (6) Comprehensive evaluation indicators Each indicator is dimensionless, that is, the autonomous average speed is divided by the designed autonomous average speed, the autonomous maximum speed is divided by the designed autonomous maximum speed, the autonomous climbing ability, the autonomous trench crossing ability, and the autonomous boss passing ability are divided by the corresponding number of tests, in order to evaluate the autonomous maneuvering function and performance of the ground unmanned vehicle.

Images