Cloud control scheduling method and system for testing autonomous vehicle, electronic device, and medium

Abstract

The application discloses an automatic driving vehicle test cloud control scheduling method and system, electronic equipment and medium, comprising: acquiring a space-time trajectory range corresponding to a test vehicle according to a test scene, a test task and a corresponding test rule; setting an optimization scheduling target with a safety distance as a constraint; solving the optimization scheduling target based on the space-time trajectory range corresponding to the test vehicle to obtain a time interval of the test vehicle entering a test site and a test start time; determining whether a motion object is in a dangerous domain and is about to collide at the current time according to the speed of each motion object in the test scene and the distance between adjacent motion objects; and if the motion object is in the dangerous domain and is about to collide at the current time, controlling the test vehicle to brake urgently to stop.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of automatic driving test of vehicles, and particularly relates to an automatic driving vehicle test cloud control scheduling method and system, an electronic device and a medium. BACKGROUND

[0002] At present, the automatic driving test field mainly adopts manual scheduling, and the following problems exist:

[0003] (1) Low efficiency of test arrangement: due to the limitation of manual scheduling, the current test site only allows one test to be carried out at the same time in space, resulting in a large waste of space-time resources. For example, two tests may cross at some time points, but the space they actually occupy is very limited. At this time, manual scheduling will not arrange the two tests at the same time, resulting in very low resource utilization.

[0004] (2) Low efficiency of test preparation: manual cleaning and arrangement of the site are required before the test, manual arrangement of in-vehicle data recording and monitoring equipment, and operation, safety training, etc. of the staff participating in the test are required, and manual cleaning of the site, restoration of the original settings of the vehicle, etc. are required after the test.

[0005] (3) Low efficiency of test evaluation: whether part of the test project is successful or not needs to be evaluated manually according to the vehicle-mounted sensor after the test, and whether part of the test is successful or not is determined by whether a collision occurs during the test. If the vehicle or other test props participating in the test are damaged due to real collision or other reasons, manual inspection and repair work is required.

[0006] These problems together cause the existing automatic driving test site to be unable to meet the demand of the test demand side for obtaining test data at low cost and high efficiency. Therefore, it is urgent to propose an automatic driving vehicle test scheduling method to overcome the above problems. SUMMARY

[0007] The present application aims to provide an automatic driving vehicle test cloud control scheduling method, system, electronic device and medium to overcome the deficiencies of the prior art.

[0008] The purpose of the present application is achieved by the following technical solutions:

[0009] In a first aspect, the present application provides an automatic driving vehicle test cloud control scheduling method, which comprises:

[0010] According to the acquired test scene, test task and corresponding test rules, the time-space trajectory range corresponding to the test vehicle is acquired;

[0011] Set an optimization scheduling target with a safety distance as a constraint; based on the time-space trajectory range corresponding to the test vehicle, solve the optimization scheduling target to obtain the time interval of the test vehicle entering the test site and the test start time.

[0012] In a second aspect, the embodiment of the present application provides an automatic driving vehicle test cloud control scheduling system, which comprises:

[0013] A planning module is configured to: acquire a time-space trajectory range corresponding to a test vehicle according to the acquired test scene, test task and corresponding test rule; set an optimization scheduling target with a safety distance as a constraint; and based on the time-space trajectory range corresponding to the test vehicle, solve the optimization scheduling target to obtain the time interval of the test vehicle entering the test site and the test start time.

[0014] A control module is configured to schedule the movement of the test vehicle according to the time interval of the test vehicle entering the test site and the test start time output by the planning module.

[0015] In a third aspect, the embodiment of the present application provides an electronic device comprising a memory and a processor, wherein the memory is coupled to the processor; the memory is configured to store program data, and the processor is configured to execute the program data to implement the automatic driving vehicle test cloud control scheduling method described above.

[0016] In a fourth aspect, the embodiment of the present application provides a computer readable storage medium having a computer program stored thereon, wherein the program is executed by a processor to implement the automatic driving vehicle test cloud control scheduling method described above.

[0017] Compared with the prior art, the present application has the following advantages:

[0018] The present application provides an automatic driving vehicle test cloud control scheduling method, which acquires a time-space trajectory range corresponding to a test vehicle according to the acquired test scene, test task and corresponding test rule. Then, an optimization scheduling target is set with a safety distance as a constraint; based on the time-space trajectory range corresponding to the test vehicle, the optimization scheduling target is solved to obtain the time interval of the test vehicle entering the test site and the test start time. The method of the present application can be applied to a wide range of test scenes, can fully utilize time-space resources, reasonably arrange the test order of user test requirements, and realize safe and efficient automatic driving test. BRIEF DESCRIPTION OF DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiment description. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without creative labor.

[0020] Figure 1 A flowchart of an automatic driving vehicle test cloud control scheduling method provided by an embodiment of the present application is shown in the figure.

[0021] Figure 2 A schematic diagram of the time interval at which the test vehicle enters the test site and the test start time provided by an embodiment of the present application is shown in the figure.

[0022] Figure 3 A schematic diagram of the target vehicle stop-go test in the double straight test provided by an embodiment of the present application is shown in the figure.

[0023] Figure 4 A schematic diagram of an automatic driving vehicle test cloud control scheduling system provided by an embodiment of the present application is shown in the figure.

[0024] Figure 5 A schematic diagram of an electronic device provided by an embodiment of the present application is shown in the figure. DETAILED DESCRIPTION

[0025] The applicant believes that, after carefully reading the application file, accurately understanding the implementation principle and application purpose of the present application, and in combination with the existing known technology, the skilled person in the art can reproduce it by using the skills mastered, so the specific content of the present application will not be described in detail.

[0026] In addition, various schematic diagrams of the present application are shown in the accompanying drawings of the present application. In order to clearly express, some details are enlarged, and some details may be omitted. The shape of each component shown in the figure and their relative size, position relationship are only exemplary.

[0027] As shown in the figure, the present application provides an automatic driving vehicle test cloud control scheduling method, which comprises: Figure 1

[0028] Step S1, according to the test scene, test task and its corresponding test rule obtained, the space-time trajectory range corresponding to the test vehicle is obtained.

[0029] Further, the test scene includes but is not limited to: straight, curve, crossroad, ramp flow, roundabout. The test scene faced by the present example is extensive, and the method of the present application can be applied to various test sites.

[0030] Further, the test task includes but is not limited to: following task, collision avoidance task.

[0031] Further, the test rule analysis process includes: obtaining the maximum and minimum speed and acceleration of the test vehicle, the reaction time of the test vehicle to other moving objects, and the motion relationship between the moving objects, obtaining the upper and lower bounds of the space-time trajectory range corresponding to the test vehicle. ​

[0032] For example, when the test task is an avoidance task, the avoidance task is to avoid a dummy crossing the road, and the corresponding test rule is that the test road is a long straight road with one-way double lanes. If the road speed limit v max ≥ 60 km / h, the initial speed of the test vehicle is 60 km / h; if the road speed limit v max < 60 km / h, the initial speed of the test vehicle is 40 km / h. The test vehicle drives in one side lane, and when the predicted collision time reaches 3.5-4.5 s, the pedestrian crosses the road at a speed of 5-6.5 km / h on the left side of the test vehicle and passes through the rightmost lane line. The requirement is that no collision occurs. The corresponding rule analysis process is that for the upper bound, the test vehicle should maintain as high a speed as possible, and the dummy should enter the road as soon as possible. Therefore, the maximum acceleration and deceleration are taken, and the maximum reaction time is taken. The test vehicle first accelerates to the maximum speed with the maximum acceleration, and drives at the maximum speed until it reacts to the dummy and decelerates with the maximum deceleration. For the lower bound, the test vehicle should maintain as low a speed as possible, and the dummy should leave the road as late as possible. Therefore, the minimum acceleration and deceleration that can complete the avoidance task are taken as the acceleration and deceleration of the vehicle, and the reaction time is taken as 0. The test vehicle drives at the initial speed until it reacts to the dummy and decelerates with the minimum deceleration.

[0033] Further, the expression of the corresponding spatiotemporal trajectory range of the test vehicle is as follows:

[0034]

[0035] In the formula, f upper represents the upper bound of the spatiotemporal trajectory range of the test vehicle, f lower represents the lower bound of the spatiotemporal trajectory range of the test vehicle, p is the spatial position of the test vehicle; t is the time; B1 is the set of all spatiotemporal points located on the upper bound of the test spatiotemporal trajectory range; and B2 is the set of all spatiotemporal points located on the lower bound of the test spatiotemporal trajectory range.

[0036] In step S2, the optimization scheduling target is set with the safety distance as the constraint, and the optimization scheduling target is solved based on the corresponding spatiotemporal trajectory range of the test vehicle to obtain the time interval of the test vehicle entering the test site and the test start time.

[0037] The expression of the optimization scheduling target is as follows:

[0038]

[0039]

[0040] In the formula, f lower The relationship between the position d1 of the moving object corresponding to the lower bound of the past test spatiotemporal trajectory range and the time t; fupper d2 is the position of the moving object corresponding to the upper bound of the spatio-temporal trajectory range of the current test; t is the relationship between the position d2 and the time t; min dmin is the minimum distance difference between the lower bound of the spatio-temporal trajectory range of the past test and the upper bound of the spatio-temporal trajectory range of the current test for each time point within the overlap time of the two tests; d safety dmin is the minimum distance difference between the lower bound of the spatio-temporal trajectory range of the past test and the upper bound of the spatio-temporal trajectory range of the current test for each time point within the overlap time of the two tests; d in t1 is the minimum time of the lower bound of the spatio-temporal trajectory range of the past test; t2 is the maximum time of the upper bound of the spatio-temporal trajectory range of the current test; t start t is the time interval between the time when the test vehicle enters the test site and the time when the test vehicle starts the test; t f t is the test start time.

[0041] It should be noted that, as Figure 2 shown, the present example reorganizes the test demand queue based on the adaptability of each test spatio-temporal range according to the user test demand, optimally sorts, and iteratively solves the optimal scheduling scheme of all test scenarios in the demand queue according to the current test total spatio-temporal range boundary, the test demand at the next moment, and the minimum safety distance condition, reasonably arranges the test order of the user test demand, and improves the efficiency of the automatic driving test.

[0042] Further, the method further comprises:

[0043] Step S3, determining whether a moving object is in a dangerous domain and about to collide at the current moment according to the speed of each moving object in the test scenario and the distance between adjacent moving objects; if a moving object is in a dangerous domain and about to collide at the current moment, controlling the test vehicle to brake urgently to stop.

[0044] Further, the expression of the dangerous domain is as follows:

[0045]

[0046] In the formula, v f is the speed of the front moving object; v r is the speed of the rear moving object; d int is the distance between the front moving object and the rear moving object; d danger is the dangerous distance; a max,de is the maximum emergency braking acceleration.

[0047] Further, the moving object includes but is not limited to: a test vehicle, a target vehicle, a pedestrian, a bicycle, etc. located in the test scenario.

[0048] It should be noted that in this example, the speed, acceleration, heading angle of the automatic driving vehicle is controlled by the controller to enter and complete the test at the space-time point of the planning output (i.e. the time interval of the test vehicle entering the test site and the test start time); if the moving object is in the danger zone, then the automatic driving test vehicle is controlled to decelerate at the maximum acceleration of the emergency brake a max,de , to ensure the safety of the test.

[0049] Embodiment 1

[0050] As shown in Figure 3 , this example takes the target vehicle stop-go test in the double straight test as an example to explain the automatic driving vehicle test cloud control scheduling method provided by the application. Specifically, it includes the following steps:

[0051] Step S1, according to the obtained test scene, test task and its corresponding test rule, obtain the scheduling state of each test, the pre-test scheduling strategy and the post-test scheduling strategy, analyze the test rule, find out the feature vector of each test and its value range, and combine the motion relationship between each moving object to obtain its space-time trajectory range; if a new test is added, the rule is reanalyzed and the space-time trajectory range is updated in real time.

[0052] Among them, the scheduling state refers to the motion state of each moving object in the test that is not in the test stage but is in the field to accept system scheduling, in this test, the speed of each vehicle in the scheduling state v sch is 10 m / s, and the distance between the test vehicle and the target vehicle in the scheduling state is 40 m.

[0053] The pre-test scheduling strategy refers to the strategy of adjusting the motion state of each moving object in the test and the motion relationship between them to the motion state required at the beginning of the test, in this test, the initial speed of the test vehicle and the target vehicle is required to be 75% v max , v max is the maximum speed, and the initial distance between the two vehicles has no requirement.

[0054] The post-test scheduling strategy refers to adjusting the motion state of each moving object and the motion relationship between them to the normal scheduling state at the end of the test. The adjustment stage in the later stage of the test is divided into a speed adjustment sub-stage and a distance adjustment sub-stage. The time length of the speed adjustment sub-stage is 5s, and the test vehicle and the target vehicle are adjusted to the scheduling speed at a uniform acceleration. The time length of the distance adjustment sub-stage depends on the actual distance between the front and rear vehicles at the end of the test. The distance adjustment strategy involves adjusting the speed of the rear vehicle according to the distance between the front and rear vehicles to ensure that the distance between the two vehicles is exactly equal to the scheduling distance. If the distance is greater than the scheduling distance, the rear vehicle first accelerates and then decelerates. If the distance is less than the scheduling distance and greater than the critical distance, the rear vehicle first decelerates and then accelerates. If the distance is less than the critical distance, the rear vehicle decelerates to a stop and then starts to accelerate again.

[0055] The speed and acceleration / deceleration of the test vehicle and the target vehicle in the later adjustment stage, and the waiting time calculation formula of the test vehicle are as follows:

[0056]

[0057]

[0058] Where, v vut is the speed of the test vehicle, v vt is the speed of the target vehicle, t is the time used after entering the later adjustment stage; d cri is the critical distance, d init is the distance between the test vehicle and the target vehicle at the end of the actual test stage; t ac,cha is the acceleration time when d init > d sch , t de,cha is the deceleration time when d init > d sch , t ac,tug is the acceleration time when d init < d sch , t de,tug is the deceleration time when d init < d sch , and t wai is the waiting time of the test vehicle when stopping.

[0059] The test has two vehicles, the test vehicle and the target vehicle, and only occupies one lane. Given v max , the variable amount of the target vehicle has an average deceleration (a vt,de ), an average acceleration (a vt,ac ), the variable amount of the test vehicle has a reaction time (t det ) from the start of deceleration of the front vehicle to the time when the front vehicle starts to decelerate and takes deceleration measures, an average deceleration (a vut,de ), and a start-up time (tact ). To reduce the freedom of the test vehicle, the test vehicle is taken over as soon as the system detects that the test vehicle has completed starting, and is accelerated at the scheduled acceleration. The feature vector of this test is extracted as:

[0060] c = (a vt,de , a vt,ac , a vut,de , t det , t act )

[0061] The expression of the spatiotemporal trajectory range corresponding to the test vehicle is as follows:

[0062]

[0063] In the formula, f upper represents the upper bound of the spatiotemporal trajectory range of the test vehicle, f lower represents the lower bound of the spatiotemporal trajectory range of the test vehicle, p is the spatial position of the test vehicle; t is the time; B1 is the set of all spatiotemporal points located on the upper bound of the test spatiotemporal trajectory range; and B2 is the set of all spatiotemporal points located on the lower bound of the test spatiotemporal trajectory range. According to the extracted feature vector, the upper and lower bounds of the out-of-control range can be analyzed and calculated in the following manner:

[0064] To calculate the upper bound of the spatiotemporal range, the target vehicle should travel at as high a speed as possible; specifically including:

[0065] First, focus on the variables a vt,de and a vt,ac in the feature vector that are related to the target vehicle: a vt,de is the deceleration of the target vehicle, so the absolute value of a vt,de should be as small as possible, so a vt,de takes the minimum value allowed by the rule; a vt,ac is the acceleration of the target vehicle, so the absolute value of a vt,ac should be as large as possible, so a vt,ac takes the maximum value a max,ac allowed by the scheduling system. In addition, since the test rule only stipulates that the target vehicle remains stationary after the speed of the test vehicle drops to 0, it does not consider the speed of the test vehicle dropping to 0 before the target vehicle decelerates to a stop, so an additional provision is made that when the speed of the test vehicle drops to 0, if the target vehicle has not stopped, the acceleration is immediately changed from deceleration to acceleration, and the acceleration requirement is consistent with the acceleration requirement of the target vehicle remaining stationary after the speed of the test vehicle drops to 0. Therefore, in order to keep the speed of the target vehicle at a high level, the test vehicle should immediately reduce its speed to 0 at the beginning of the test, i.e., when t det = 0, a vut,de , a max,deThe maximum value is taken, and then the target vehicle re-accelerates to the dispatch speed, and should maintain the dispatch speed after re-acceleration.

[0066] The lower bound of the space-time range is calculated, and the test vehicle should maintain a lower speed as much as possible; specifically including:

[0067] First, pay attention to the variables a vut,de , t det and t act in the feature vector related to the test vehicle: a vut,de is the deceleration of the test vehicle, so the absolute value of a vut,de should be as large as possible, so a vut,de takes the maximum value a max,de allowed by the dispatch system; t det is the reaction time of the test vehicle, since the subsequent action is emergency braking, t det should be as small as possible, so t det = 0; t act is the start-up time of the test vehicle, since the subsequent action is acceleration, t act should be as large as possible, so t act takes the maximum value allowed by the rules and the dispatch system. In addition, since the test vehicle takes aggressive deceleration measures during the actual test phase, it is far apart from the target vehicle, so the test vehicle will accelerate to adjust the distance between the two vehicles during the later adjustment phase. Therefore, in order to reduce the acceleration time and acceleration distance of the test vehicle during the later adjustment phase, the target vehicle should also travel at a smaller speed as much as possible to alleviate the increase in distance between the two, so a vt,de should take the maximum value allowed by the rules, and a vt,ac should take the minimum value allowed by the rules.

[0068] Step S2, set the optimization dispatch target with the safety distance as the constraint; based on the space-time trajectory range corresponding to the test vehicle, solve the optimization dispatch target to obtain the time interval of the test vehicle entering the test site and the test start time.

[0069] Specifically, first generate the space-time trajectory range of the first test with the initial parameters and the first item of the demand queue, i.e. the t in and t start or dmin of the first series of tests are the initial default values. The design idea of the initialization parameters of the first series of tests is that when each moving object of the test is converted from the test state of the last test of the entire test to the dispatch state, each moving object is in the test state, i.e. the initialization of each to-be-solved parameter is:

[0070]

[0071] or

[0072]

[0073] where: t test,i is the total time length of the i-th test in the series of tests (m is the number of tests in the series of tests); d sou,i is the absolute position of the i-th test site.

[0074] According to the spatio-temporal trajectory range of the initialization test and the initialized upper bound of the spatio-temporal trajectory, an optimization calculation is performed to obtain the next test t in and t start or d min ; and the lower bound of the spatio-temporal trajectory range is calculated, and finally the upper bound is updated; the above steps are repeated until the optimized t in and t start or d min of all tests are obtained.

[0075] The expression of the optimization objective is as follows:

[0076]

[0077]

[0078] In the formula, f lower is the relationship between the position d1 of the moving object corresponding to the lower bound of the spatio-temporal trajectory range of the past test and the time t; f upper is the relationship between the position d2 of the moving object corresponding to the upper bound of the spatio-temporal trajectory range of the current test and the time t; d min represents the minimum distance difference of each time point within the overlapping time of the lower bound of the spatio-temporal trajectory range of the past test and the upper bound of the spatio-temporal trajectory range of the current test; d safety represents the safety distance; t1 is the minimum time of the lower bound of the spatio-temporal trajectory range of the past test; t2 is the maximum time of the upper bound of the spatio-temporal trajectory range of the current test; t in represents the time interval of the test vehicle entering the test site; t start represents the test start time.

[0079] The optimization problem is essentially to find the minimum value of a constrained nonlinear multivariable function, and its solving method can be to first block multiple variables of the objective function, and then use an iterative algorithm such as Newton method, gradient descent method, etc. to iteratively solve the function of each variable, and finally the optimal solution obtained is the optimal solution of the entire objective function.

[0080] In step S3, whether a moving object is in a dangerous domain to be collided is determined according to the speed of each moving object in the test scene and the distance between adjacent moving objects.

[0081] If a moving object is in the danger zone at the current time and is about to collide, the control test vehicle is braked to stop, and the time-space trajectory of the vehicle is controlled within the test time-space trajectory range.

[0082] Specifically, taking a double straight road scene as an example, its danger zone S danger is:

[0083]

[0084] In the formula: v f is the speed of the front moving object, v r is the speed of the rear moving object, d int is the distance between the front moving object and the rear moving object, d danger is the dangerous distance, which is also a reaction index of the system's sensitivity to collision, a max,de is the maximum emergency braking acceleration specified by the system.

[0085] According to the principle of yielding speed and not yielding lane, the anti-collision intervention strategy of the double straight road scene is limited to emergency braking, and the test vehicle does not perform lane changing operation. The intervention strategy specifically shows that it is immediately braked to stop, and then waits for the actual test phase to end.

[0086] As Figure 4 shown, the present example also provides an automatic driving vehicle test cloud control scheduling system, the system comprises:

[0087] A planning module acquires the time-space trajectory range of the test vehicle according to the acquired test scene, test task and its corresponding test rule; sets an optimization scheduling target with safety distance as a constraint; solves the optimization scheduling target based on the time-space trajectory range of the test vehicle to obtain the time interval of the test vehicle entering the test site and the test start time;

[0088] A control module is configured to schedule the movement of the test vehicle according to the time interval of the test vehicle entering the test site and the test start time output by the planning module.

[0089] Specifically, the control module controls the speed, acceleration and heading angle of the automatic driving vehicle through a controller to enter and complete the test at the planning output time-space point (i.e. the time interval of the test vehicle entering the test site and the test start time); if a moving object is in the danger zone at the current time, the automatic driving test vehicle is controlled to decelerate at the maximum emergency braking acceleration, thereby ensuring the safety of the test.

[0090] The monitoring module is configured to determine whether a moving object is in a dangerous area and is about to collide with another moving object according to the speed of each moving object in the test scene and the distance between adjacent moving objects; and if a moving object is in a dangerous area and is about to collide with another moving object at the current time, the test vehicle is controlled to stop by emergency braking.

[0091] Further, the monitoring module comprises a visual sensor in the test field and vehicle sensors (including millimeter wave radar, laser radar, visual sensor, integrated inertial navigation, etc.) carried on the autonomous vehicle.

[0092] Further, in the present example, the communication between the planning module, the monitoring module and the control module adopts Ethernet communication and vehicle chassis CAN network communication.

[0093] As to the system in the above-described embodiments, the specific manner in which each module performs operations has been described in detail in the embodiments related to the method, and will not be described in detail here.

[0094] As to the system embodiments, since they basically correspond to the method embodiments, the related parts are described in the part of the method embodiments. The above-described system embodiments are only illustrative, wherein the units described as separate components can or can not be physically separated, and the components displayed as units can or can not be physical units, i.e., they can be located in one place or distributed on multiple network units. Some or all of the modules can be selected to achieve the purpose of the present application according to actual needs. Those skilled in the art can understand and implement it without creative labor.

[0095] Correspondingly, the present application also provides an electronic device, comprising: one or more processors; a memory for storing one or more programs; when the one or more programs are executed by the one or more processors, the one or more processors implement the automatic driving vehicle test cloud control scheduling method as described above. As Figure 5 shown, a hardware structure diagram of the automatic driving vehicle test cloud control scheduling method provided by the embodiment of the present application in any device with data processing capability. In addition to the processor, memory and network interface shown in the embodiment, any device with data processing capability in which the device is located usually according to the actual function of the device with data processing capability, can also include other hardware, which will not be described here. Figure 5 shown, a hardware structure diagram of the automatic driving vehicle test cloud control scheduling method provided by the embodiment of the present application in any device with data processing capability. In addition to the processor, memory and network interface shown in the embodiment, any device with data processing capability in which the device is located usually according to the actual function of the device with data processing capability, can also include other hardware, which will not be described here.

[0096] Correspondingly, the application further provides a computer readable storage medium, which stores computer instructions, and the instructions are executed by a processor to implement the automatic driving vehicle test cloud control scheduling method as described above. The computer readable storage medium can be an internal storage unit of any device with data processing capability, such as a hard disk or a memory. The computer readable storage medium can also be an external storage device, such as a plug-in hard disk, a smart media card (SMC), an SD card, a flash card, etc. Further, the computer readable storage medium can include both the internal storage unit of any device with data processing capability and the external storage device. The computer readable storage medium is used to store the computer program and other programs and data required by the device with data processing capability, and can also be used to temporarily store data that has been output or will be output.

[0097] The above only describes the preferred embodiments of the present application and is not intended to limit the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A cloud-based scheduling method for testing autonomous vehicles, characterized in that, The method includes: Based on the obtained test scenario, test task and its corresponding test rules, obtain the spatiotemporal trajectory range of the test vehicle; With a safety distance as a constraint, an optimization scheduling objective is set; based on the spatiotemporal trajectory range of the test vehicle, the optimization scheduling objective is solved to obtain the time interval between test vehicles entering the test site and the test start time; The method further includes: Based on the speed of each moving object in the test scenario and the distance between adjacent moving objects, determine whether there are any moving objects in the danger zone that are about to collide at the current moment. If a moving object is in the danger zone and is about to collide with the test vehicle, the test vehicle will be brought to an emergency stop. The expression for the danger domain is as follows: In the formula, v f v is the velocity of the moving object in front; r d represents the velocity of the moving object behind; int d represents the distance between the moving object in front and the moving object behind; danger Dangerous distance; a max,de This is the maximum acceleration during emergency braking.

2. The cloud-based control and scheduling method for testing autonomous vehicles according to claim 1, characterized in that, The test scenarios include straight roads, curves, intersections, ramp merging, and roundabouts; the test tasks include following other vehicles and collision avoidance tasks.

3. The cloud-based control and scheduling method for testing autonomous vehicles according to claim 1, characterized in that, The expression for the spatiotemporal trajectory range corresponding to the test vehicle is as follows: In the formula, f upper f represents the upper bound of the spatiotemporal trajectory range of the test vehicle. lower B1 represents the lower bound of the test vehicle's spatiotemporal trajectory range, where p is the spatial position of the test vehicle, t is the time, B1 is the set of all spatiotemporal points located at the upper bound of the test spatiotemporal trajectory range, and B2 is the set of all spatiotemporal points located at the lower bound of the test spatiotemporal trajectory range.

4. The cloud-based control and scheduling method for testing autonomous vehicles according to claim 1, characterized in that, The expression for the optimized scheduling objective is as follows: In the formula, f lower The relationship between the position d1 of the moving object and time t corresponding to the lower bound of the spatiotemporal trajectory range tested in the past; f upper This represents the relationship between the position d2 of the moving object corresponding to the upper bound of the current test's spatiotemporal trajectory range and time t; d min d represents the minimum distance difference between the lower bound of the spatiotemporal trajectory range of the past test and the upper bound of the spatiotemporal trajectory range of the current test at every time point within the overlap time of the two tests; safety t1 represents the safety distance; t1 is the minimum time of the lower bound of the spatiotemporal trajectory range in past tests; t2 is the maximum time of the upper bound of the spatiotemporal trajectory range in the current test. t in Indicates the time interval between the arrival of the test vehicle at the test site; t start Indicates the test start time.

5. A cloud-based control and scheduling system for testing autonomous vehicles, characterized in that, The system includes: The planning module obtains the spatiotemporal trajectory range of the test vehicle based on the acquired test scenario, test task and its corresponding test rules; sets the optimization scheduling objective with safety distance as constraint; and solves the optimization scheduling objective based on the spatiotemporal trajectory range of the test vehicle to obtain the time interval between test vehicles entering the test site and the test start time. The control module is used to schedule the movement of test vehicles based on the time intervals between the test vehicles entering the test site and the test start time output by the planning module. The system also includes: The monitoring module is used to determine whether there is a moving object in the danger zone that is about to collide with the test object at the current moment based on the speed of each moving object in the test scenario and the distance between adjacent moving objects; if there is a moving object in the danger zone that is about to collide with the test object at the current moment, the test vehicle is controlled to brake urgently to a stop. The expression for the danger domain is as follows: In the formula, v f v is the velocity of the moving object in front; r d represents the velocity of the moving object behind; int d represents the distance between the moving object in front and the moving object behind; danger Dangerous distance; a max,de This is the maximum acceleration during emergency braking.

6. An electronic device comprising a memory and a processor, characterized in that, The memory is coupled to the processor; wherein the memory is used to store program data, and the processor is used to execute the program data to implement the autonomous vehicle test cloud control scheduling method according to any one of claims 1-4.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the cloud control scheduling method for testing autonomous vehicles as described in any one of claims 1-4.

Images