Unmanned vehicle start-stop control method and device and unmanned vehicle

Abstract

The invention provides an unmanned vehicle start-stop control method and device and an unmanned vehicle, and relates to the technical field of unmanned vehicle control. The method comprises the following steps: when the unmanned vehicle enters a parking state, detecting whether the unmanned vehicle meets a start-stop protection condition or not; wherein the start-stop protection condition is related to one or more of the following factors: a parking reason of the unmanned vehicle; the current speed of the unmanned vehicle; the current track attribute of the unmanned vehicle; external entity behavior characteristics; performing upstream false detection on a frame loss condition; the real-time accuracy of self-vehicle identification is realized; under the condition that the unmanned vehicle meets the start-stop protection condition, the unmanned vehicle is controlled to enter a start-stop protection mode, and an emergency stop instruction is issued; wherein the emergency stop instruction is used for indicating the unmanned vehicle to execute emergency stop operation. According to the embodiment of the invention, the starting and stopping times of the unmanned vehicle can be reduced by issuing the emergency stopping instruction after the specific factor is detected, so that the loss of the unmanned vehicle on hardware and energy consumption is reduced, and the driving safety is improved.

Description

Technical Field

[0001] The present disclosure relates to the field of unmanned vehicle control, and in particular to an unmanned vehicle start-stop control method and device, and an unmanned vehicle. Background Art

[0002] In autonomous driving scenarios, vehicles often face the problem of frequent starts and stops. Frequent starts and stops can cause vehicle hardware wear and tear, increase energy consumption, and potentially lead to incorrect decisions by other vehicles, posing safety risks. Summary of the Invention

[0003] To overcome the problems existing in the related art, the present disclosure provides a start-stop control method and device for an unmanned vehicle, and an unmanned vehicle.

[0004] According to a first aspect of an embodiment of the present disclosure, a method for controlling the start and stop of an unmanned vehicle is provided, comprising: When the unmanned vehicle enters a parking state, detecting whether the unmanned vehicle meets the start-stop protection conditions; wherein the start-stop protection conditions are related to one or more of the following factors: the reason for the unmanned vehicle to stop; the current speed of the unmanned vehicle; the current trajectory attributes of the unmanned vehicle; the behavior characteristics of external entities; the upstream false detection frame loss situation; the real-time accuracy of the self-vehicle identification; When the unmanned vehicle meets the start-stop protection condition, the unmanned vehicle is controlled to enter the start-stop protection mode and an emergency stop instruction is issued; wherein, the emergency stop instruction is used to instruct the unmanned vehicle to perform an emergency stop operation.

[0005] In some embodiments, the method further comprises: During the emergency stop operation, the unmanned vehicle is detected to determine whether it meets the start-stop protection exit condition; wherein the start-stop protection exit condition is related to one or more of the following factors: the current trajectory attributes of the unmanned vehicle; the number of times the unmanned vehicle continuously outputs the starting trajectory; the accumulated parking time of the unmanned vehicle; the reason for the unmanned vehicle's parking; the behavioral characteristics of external entities; the upstream false detection frame loss situation; and the real-time accuracy of self-vehicle identification. When the unmanned vehicle meets the start-stop protection exit condition, the unmanned vehicle is controlled to exit the start-stop protection mode.

[0006] In some embodiments, during the execution of the emergency stop operation, detecting whether the unmanned vehicle meets the start-stop protection exit condition includes: During the emergency stop operation, obtaining trajectory parameters of the unmanned vehicle, the trajectory parameters including: current trajectory attributes of the unmanned vehicle and the number of times the unmanned vehicle continuously outputs a starting trajectory; Based on the trajectory parameters, determining whether the unmanned vehicle meets the start-stop protection exit condition; When the unmanned vehicle satisfies the start-stop protection exit condition, controlling the unmanned vehicle to exit the start-stop protection mode includes: When the current trajectory of the unmanned vehicle is a starting trajectory and the number of times the unmanned vehicle continuously outputs the starting trajectory reaches a preset number, controlling the unmanned vehicle to exit the start-stop protection mode; The starting trajectory is a driving trajectory in which the speed parameter of at least some path points is greater than zero.

[0007] In some embodiments, the method further comprises: If the current trajectory of the unmanned vehicle is a starting trajectory and the number of times the unmanned vehicle continuously outputs the starting trajectory does not reach a preset number, continue to issue an emergency stop instruction; and / or When the current trajectory of the unmanned vehicle is not a starting trajectory, the emergency stop instruction continues to be issued.

[0008] In some embodiments, when the current trajectory of the unmanned vehicle is a starting trajectory and the number of times the unmanned vehicle continuously outputs the starting trajectory reaches a preset number, controlling the unmanned vehicle to exit the start-stop protection mode includes: Initializing a counter, the counter being used to record the number of times the unmanned vehicle continuously outputs the starting trajectory; Repeat the following steps until the count in the counter is equal to a preset threshold, and then control the unmanned vehicle to exit the start-stop protection mode: When the trajectory output by the unmanned vehicle is a starting trajectory, the count in the counter is increased by one; When the trajectory output by the unmanned vehicle is a non-starting trajectory, the count in the counter is cleared.

[0009] In some embodiments, obtaining the trajectory parameters of the unmanned vehicle includes: Periodically acquiring trajectory parameters of the unmanned vehicle according to a preset frame interval; The determining, based on the trajectory parameters, whether the unmanned vehicle satisfies a start-stop protection exit condition includes: After each completion of trajectory parameter acquisition, determining whether the unmanned vehicle meets the start-stop protection exit condition based on the acquired trajectory parameters; When the unmanned vehicle does not meet the start-stop protection exit condition, the emergency stop instruction continues to be issued.

[0010] In some embodiments, during the execution of the emergency stop operation, detecting whether the unmanned vehicle meets the start-stop protection exit condition includes: During the emergency stop operation, the unmanned vehicle is detected to determine whether it meets one or more of the following conditions: The current trajectory of the unmanned vehicle is the starting trajectory; The number of times the unmanned vehicle continuously outputs the starting trajectory reaches a preset number; The accumulated parking time of the unmanned vehicle reaches a preset time; The unmanned vehicle fails to detect a preset parking reason; The actions of external entities do not interfere with the driving trajectory of the autonomous vehicle; The upstream false detection rate of the unmanned vehicle is less than a preset false detection rate threshold; The number of consecutive frame losses of the unmanned vehicle is less than a preset frame threshold; The self-vehicle recognition error of the unmanned vehicle is less than a preset error threshold.

[0011] In some embodiments, when the unmanned vehicle enters a parking state, detecting whether the unmanned vehicle meets a start-stop protection condition includes: When the unmanned vehicle enters the parking state, it is detected whether the unmanned vehicle meets one or more of the following conditions: The unmanned vehicle enters a parking state due to a preset parking reason; The current speed of the unmanned vehicle is less than a preset speed threshold; The current trajectory of the unmanned vehicle is a non-starting trajectory; The behavior of an external entity interferes with the driving trajectory of the autonomous vehicle; The upstream false detection rate of the unmanned vehicle is greater than a preset false detection rate threshold; The number of consecutive frame drops of the unmanned vehicle is greater than a preset frame threshold; The self-vehicle recognition error of the unmanned vehicle is greater than a preset error threshold.

[0012] According to a second aspect of the present disclosure, a start-stop control device for an unmanned vehicle is provided, comprising: A detection module is configured to detect whether the unmanned vehicle meets the start-stop protection conditions when the unmanned vehicle enters a parking state; wherein the start-stop protection conditions are related to one or more of the following factors: the reason for the unmanned vehicle to stop; the current speed of the unmanned vehicle; the current trajectory attributes of the unmanned vehicle; the behavioral characteristics of external entities; the upstream false detection frame loss situation; and the real-time accuracy of self-vehicle identification. A control module is used to control the unmanned vehicle to enter a start-stop protection mode and issue an emergency stop instruction when the unmanned vehicle meets the start-stop protection conditions; wherein the emergency stop instruction is used to instruct the unmanned vehicle to perform an emergency stop operation.

[0013] According to a third aspect of the present disclosure, an unmanned vehicle is provided, wherein the unmanned vehicle is used to execute the method described in the first aspect.

[0014] The solution provided by the embodiment of the present disclosure detects whether the unmanned vehicle meets the start-stop protection conditions when the unmanned vehicle enters the parking state. The start-stop protection conditions are related to one or more of the following factors: the reason for the unmanned vehicle to stop; the current speed of the unmanned vehicle; the current trajectory attributes of the unmanned vehicle; the behavioral characteristics of external entities; the upstream false detection and frame loss situation; the real-time accuracy of the self-vehicle identification. When the unmanned vehicle meets the start-stop protection conditions, the unmanned vehicle is controlled to enter the start-stop protection mode and an emergency stop instruction is issued; wherein the emergency stop instruction is used to instruct the unmanned vehicle to perform an emergency stop operation. The embodiment of the present disclosure can reduce the number of starts and stops of the unmanned vehicle by issuing an emergency stop instruction after detecting specific factors, thereby reducing the loss of hardware and energy consumption of the unmanned vehicle and improving driving safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 A flow chart of a method for controlling the start and stop of an unmanned vehicle in an embodiment of the present disclosure is shown.

[0016] Figure 2 A flow chart showing another method for controlling the start and stop of an unmanned vehicle in an embodiment of the present disclosure is shown.

[0017] Figure 3 A schematic structural diagram of a start-stop control device for an unmanned vehicle in an embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0018] Exemplary embodiments will be described in detail herein, with examples illustrated in the accompanying drawings. In the following description, when referring to the drawings, identical numerals in different figures represent identical or similar elements, unless otherwise indicated. The embodiments described in the following exemplary embodiments are not intended to represent all possible embodiments consistent with the present disclosure. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present disclosure, as detailed in the appended claims.

[0019] In addition, the terms “first”, “second”, etc. used in the present disclosure are only used to distinguish and describe, and cannot be understood as indicating or implying relative importance.

[0020] First, an embodiment of the present disclosure provides a method for controlling the start and stop of an unmanned vehicle, which can be executed by any unmanned vehicle.

[0021] Figure 1 A schematic diagram of a start-stop control method for an unmanned vehicle according to an embodiment of the present disclosure is shown. Figure 1 As shown, the method includes the following S101 to S102.

[0022] S101, when the unmanned vehicle enters a parking state, detecting whether the unmanned vehicle meets the start-stop protection conditions.

[0023] S102: When the unmanned vehicle meets the start-stop protection conditions, the unmanned vehicle is controlled to enter the start-stop protection mode and an emergency stop command is issued.

[0024] In some embodiments, the start-stop protection conditions are related to one or more of the following factors: the reason for the unmanned vehicle to stop; the current speed of the unmanned vehicle; the current trajectory attributes of the unmanned vehicle; the behavioral characteristics of external entities; the upstream false detection and frame loss situation; and the real-time accuracy of self-vehicle identification.

[0025] It's understandable that the purpose of putting the autonomous vehicle into start-stop protection mode is to prevent it from frequently starting and stopping within a short period of time. The reason for including the above factors as factors related to the start-stop protection conditions is that they can be used to indicate whether the autonomous vehicle needs to stop and whether the reason for stopping is likely to have a lasting impact on the autonomous vehicle, thereby causing the autonomous vehicle to frequently start and stop within a short period of time. The following will detail the correlation between the above factors and the start-stop protection conditions.

[0026] For example, the start-stop protection condition can be associated with the reason for the driverless vehicle to stop. For example, suppose the driverless vehicle enters a stop state due to a traffic light change while driving. This is generally a normal and expected stopping behavior, so the driverless vehicle does not need to enter the start-stop protection mode. However, if the driverless vehicle enters a stop state due to an emergency or yielding to another obstacle while driving, the driverless vehicle can enter the start-stop protection mode to prevent the driverless vehicle from frequently starting and stopping due to the continuous instability caused by the emergency. Among them, the emergency situation can be an unforeseen situation such as the sudden stop of the vehicle in front or the sudden approach of an external moving obstacle. Therefore, whether the driverless vehicle enters the stop state due to a preset reason can be used as one of the factors associated with the start-stop protection condition.

[0027] For example, the start-stop protection condition can be associated with the current speed of the unmanned vehicle. It is worth noting that although S101 mentions detecting whether the unmanned vehicle meets the start-stop protection condition when the unmanned vehicle enters the "parking state", the "parking state" here does not mean that the unmanned vehicle is in a completely stationary state, but is used to indicate the state after the decision-making system of the unmanned vehicle decides to issue a parking command. Although the speed of the unmanned vehicle in the "parking state" will gradually decrease, before the unmanned vehicle comes to a complete stop, the external factors that caused the decision-making system to issue the parking command may have been eliminated, and the unmanned vehicle will re-enter the normal driving state. In other words, the unmanned vehicle has restarted before it comes to a complete stop. This situation does not fall into the category of frequent start-stop and the damage to the unmanned vehicle is relatively small, so there is no need to enter the start-stop protection mode.

[0028] Based on this, the current speed of the autonomous vehicle is considered one of the factors associated with the start-stop protection condition. For example, a preset speed threshold (e.g., 0.1 m / s) can be configured. When the autonomous vehicle's current speed falls below the preset speed threshold, the autonomous vehicle is considered stationary. This can be combined with other start-stop protection conditions to control whether the autonomous vehicle enters the start-stop protection mode.

[0029] For example, the start-stop protection condition can be associated with the current trajectory attributes of the unmanned vehicle. It should be noted that after the unmanned vehicle enters the parking state, the trajectory attributes of its subsequent decision output can be roughly divided into two categories: the starting trajectory and the non-starting trajectory (parking trajectory). The starting trajectory can be understood as a driving trajectory in which the speed parameter of at least some path points is greater than zero, i.e., a driving trajectory used to instruct the unmanned vehicle to start running; the non-starting trajectory can be understood as a driving trajectory in which the speed parameter of each path point is equal to zero, i.e., a driving trajectory used to keep the unmanned vehicle stationary.

[0030] If the autonomous vehicle's current trajectory is a parking trajectory, it means the vehicle should have stopped, potentially triggering the start-stop protection mode. If the vehicle's current trajectory is a non-starting trajectory, it means the vehicle is not coming to a complete stop, and there's no need to enter the start-stop protection mode. Therefore, the vehicle's current trajectory can be used as one of the factors associated with the start-stop protection condition.

[0031] For example, the start-stop protection condition can be associated with the behavioral characteristics of external entities. These external entities can be obstacles such as vehicles and pedestrians. If an external entity interferes with the vehicle's currently planned trajectory while the autonomous vehicle is driving, the vehicle may be forced to stop due to the external entity's behavior. Because the behavior of external entities is uncertain, the autonomous vehicle can enter a start-stop protection mode to avoid frequent starts and stops caused by external entities.

[0032] For example, the same pedestrian or vehicle might repeatedly cross in front of the autonomous vehicle, causing a persistent impact for a certain period of time. Without start-stop protection mode, the autonomous vehicle might stop each time the pedestrian or vehicle returns to the front of the vehicle and start each time the pedestrian or vehicle leaves the front of the vehicle, resulting in frequent starts and stops.

[0033] For example, suppose the interference of an external entity on the autonomous vehicle's trajectory is brief and non-continuous. In this case, even if the autonomous vehicle starts immediately after detecting that the interference has been resolved, it may not stop again due to the same external entity. However, due to the uncertainty of the external entity's behavior, it is impossible to predict whether it will suddenly change its current behavior. Moreover, there may be multiple external entities that combine to cause a near-continuous interference effect on the autonomous vehicle (for example, multiple pedestrians crossing in front of the autonomous vehicle one by one). Therefore, entering the start-stop protection mode in this case still makes sense.

[0034] Therefore, whether the external entity interferes with the driving trajectory of the unmanned vehicle can be used as one of the factors associated with the start-stop protection condition.

[0035] For example, the start-stop protection condition can be associated with the upstream false detection and frame loss of the unmanned vehicle, which can be described here as the false detection rate and the number of frame losses.

[0036] Regarding false detections, autonomous vehicles rely on various sensors (such as lidar and cameras) to detect conditions around them while driving. Due to uneven light reflection or weather conditions (such as fog or strong sunlight), these sensors may cause false detections, leading to inaccurate recognition of external entities. If the upstream false detection rate exceeds a preset false detection rate threshold, the vehicle's current perception data is deemed unreliable. To mitigate driving risks, the vehicle can enter a parking state and resume driving when the upstream false detection rate is less than or equal to the preset false detection rate threshold. To prevent frequent starts and stops caused by fluctuations in the upstream false detection rate, the vehicle can enter a start-stop protection mode. Therefore, the upstream false detection rate can be used as a factor associated with the start-stop protection condition.

[0037] Regarding frame loss, autonomous vehicles use sensors to continuously capture data frames (e.g., image frames, point cloud frames) of their surroundings while driving. If, under certain circumstances, the sensors are unable to collect valid data frames due to weather conditions (such as heavy snow or rain), or if the autonomous vehicle's perception system malfunctions, frame loss may occur. If excessive frame loss occurs continuously (greater than a preset frame count threshold), the autonomous vehicle may misjudge its surroundings due to incomplete perception information. In this case, to reduce the risk of autonomous vehicle driving, the autonomous vehicle can enter a parking state and resume driving after the frame loss situation improves. To prevent the autonomous vehicle from resuming driving prematurely before the frame loss situation has completely resolved, the frame loss situation can be used as a factor associated with the start-stop protection condition to prevent the autonomous vehicle from frequently starting and stopping due to fluctuations in frame loss, which could lead to unnecessary driving behavior.

[0038] For example, the start-stop protection condition can be linked to the driverless vehicle's self-vehicle recognition accuracy. While driving, the driverless vehicle can use sensor-based perception algorithms to perceive its own position in real time and coordinate with the surrounding environment. However, in certain special circumstances, such as when the lidar or camera sensors are obscured, or when road signs and the surrounding environment become unclear, the system may experience significant self-vehicle recognition errors. When the self-vehicle recognition error exceeds a preset error threshold, it can be determined that the vehicle's position cannot be accurately determined, triggering a stop command to ensure a safe stop and avoid accidents. To prevent the driverless vehicle from frequently starting and stopping due to fluctuations in self-vehicle recognition error, the vehicle can be put into a start-stop protection mode. Therefore, self-vehicle recognition accuracy can be included as one of the factors associated with the start-stop protection condition.

[0039] It is worth noting that the above factors can be used alone or in combination when used to construct the start-stop protection conditions, and the embodiments of the present disclosure do not limit this.

[0040] In combination with the above examples, it can be seen that the start-stop protection conditions in the above S101 can be specifically described as, when the unmanned vehicle enters the parking state, detecting whether the unmanned vehicle meets one or more of the following conditions: the unmanned vehicle enters the parking state due to a preset parking reason; the current speed of the unmanned vehicle is less than the preset speed threshold; the current trajectory of the unmanned vehicle is a non-starting trajectory; the behavior of the external entity interferes with the driving trajectory of the unmanned vehicle; the upstream false detection rate of the unmanned vehicle is greater than the preset false detection rate threshold; the number of consecutive frame losses of the unmanned vehicle is greater than the preset frame number threshold; the self-vehicle recognition error of the unmanned vehicle is greater than the preset error threshold.

[0041] For example, in practical applications, three conditions can be selected as start-stop protection conditions for the unmanned vehicle: the unmanned vehicle entering a stopped state due to a preset stopping reason, the unmanned vehicle's current speed being less than a preset speed threshold, and the unmanned vehicle's current trajectory being a non-starting trajectory. That is, when the unmanned vehicle meets all three conditions simultaneously, the unmanned vehicle is controlled to enter start-stop protection mode. The combination of these three conditions can minimize frequent starts and stops for the unmanned vehicle while ensuring its efficient passage.

[0042] In some embodiments, an emergency stop command is used to instruct the unmanned vehicle to perform an emergency stop. It is understood that the emergency stop command has a higher priority in the unmanned vehicle's control system. Even if the decision-making system outputs a starting trajectory based on the unmanned vehicle's current perception data, intending the unmanned vehicle to enter a driving state, the presence of the emergency stop command will cause the unmanned vehicle to remain stationary.

[0043] That is to say, when the unmanned vehicle enters the start-stop protection mode, the unmanned vehicle will be forced to remain stationary for a certain period of time under the action of the emergency stop command to avoid frequent starting and stopping of the unmanned vehicle in a short period of time.

[0044] For example, when the unmanned vehicle is in the start-stop protection mode, the unmanned vehicle can send an emergency stop command to the instruction execution module after each decision-making system outputs a driving trajectory, so that the unmanned vehicle in the start-stop protection mode will not be affected by the decision trajectory and start prematurely.

[0045] The above embodiments provide a detailed description of the start-stop control method for unmanned vehicles provided by the present disclosure. Based on this, the present disclosure can accurately determine when an unmanned vehicle enters start-stop protection mode by setting relatively reasonable start-stop protection conditions. If the unmanned vehicle meets the start-stop protection conditions, the vehicle is controlled to enter start-stop protection mode and an emergency stop command is issued to maintain the vehicle in a stopped state, thus reducing unnecessary start-stop behavior.

[0046] Next, we will combine Figure 2 The method of exiting the start-stop protection mode in the embodiment of the present disclosure is described.

[0047] Please refer to Figure 2 , Figure 2 A flow chart showing another method for controlling the start and stop of an unmanned vehicle according to an embodiment of the present disclosure is shown. Figure 2 As shown, the method includes the following S201 to S202.

[0048] S201, during the emergency stop operation, detecting whether the unmanned vehicle meets the start-stop protection exit conditions.

[0049] S202 , when the unmanned vehicle meets the start-stop protection exit conditions, control the unmanned vehicle to exit the start-stop protection mode.

[0050] Among them, the start-stop protection exit conditions are related to one or more of the following factors: the current trajectory attributes of the unmanned vehicle; the number of times the unmanned vehicle continuously outputs the starting trajectory; the cumulative parking time of the unmanned vehicle; the reason for the unmanned vehicle to stop; the behavioral characteristics of the external entity; the upstream false detection and frame loss situation; and the real-time accuracy of self-vehicle identification.

[0051] In some embodiments, the start-stop protection exit condition is intended to characterize whether the internal factors and / or external factors that cause the unmanned vehicle to enter the start-stop protection mode have been eliminated, and whether the unmanned vehicle can resume normal operation. Therefore, the setting of the start-stop protection exit condition of the unmanned vehicle has certain similarities with the setting of the aforementioned start-stop protection condition.

[0052] Next, the correlation between the above factors and the start-stop protection exit conditions will be introduced in detail.

[0053] For example, the start-stop protection exit condition can be associated with the vehicle's current trajectory attributes. As previously mentioned, after the vehicle enters a parked state, its subsequent trajectory attributes can be broadly divided into two categories: a starting trajectory and a non-starting trajectory (parking trajectory). If the vehicle's decision-making system outputs a starting trajectory, it can be determined that the vehicle meets the starting conditions, and this can be used as one of the factors associated with the start-stop protection exit condition.

[0054] For example, the start-stop protection exit condition can be linked to the number of times the unmanned vehicle continuously outputs a starting trajectory. Each time the unmanned vehicle outputs a starting trajectory, it means that the vehicle meets the starting conditions at the moment the starting trajectory is output. However, it is unpredictable whether the trajectory output by the unmanned vehicle will change to a non-starting trajectory the next moment due to interference. Therefore, if the unmanned vehicle only outputs a starting trajectory once and then exits the start-stop protection mode and resumes normal driving, it may stop again due to the non-starting trajectory output by the unmanned vehicle the next moment. On the other hand, when the unmanned vehicle continuously outputs the starting trajectory, it means that the vehicle meets the starting conditions during the continuous period of outputting the starting trajectory. In this case, although the attributes of the trajectory that the unmanned vehicle decides to output at the next moment are still unknown, the fact that the unmanned vehicle has continuously output the starting trajectory indicates that the environment around the unmanned vehicle has stabilized, and there is a high probability that the unmanned vehicle will continue to output the starting trajectory in the next moment. Therefore, the number of times the unmanned vehicle continuously outputs the starting trajectory can be used as one of the factors associated with the start-stop protection exit condition.

[0055] For example, the start-stop protection exit condition can be linked to the accumulated parking time of the unmanned vehicle. On the one hand, the interference factors that cause the unmanned vehicle to stop and enter the start-stop protection mode are often persistent and recurring, and some interference factors may disappear over time. Therefore, by keeping the unmanned vehicle parked continuously, the unmanned vehicle can wait in place for the interference factor to disappear. When the accumulated parking time of the unmanned vehicle reaches a preset time, it is determined that there is a high probability that the interference factor has been eliminated. Therefore, the accumulated parking time of the unmanned vehicle can be used as one of the factors associated with the start-stop protection exit condition.

[0056] For example, the start-stop protection exit condition can be associated with the reason for the unmanned vehicle's parking. As described above in the start-stop protection condition, upon detecting a preset parking reason, the unmanned vehicle may enter the start-stop protection mode. Accordingly, when setting the start-stop protection exit condition, it can be configured to ensure that the preset parking reason that caused the unmanned vehicle to enter the start-stop protection mode has been eliminated.

[0057] For example, the start-stop protection exit condition can be associated with the behavior of an external entity. If the external entity's behavior interferes with the autonomous vehicle's driving trajectory, even if the autonomous vehicle meets the starting conditions, it may stop again after starting due to the external entity's behavior. Therefore, the interference of the external entity with the autonomous vehicle's driving trajectory can be used as one of the factors associated with the start-stop protection exit condition.

[0058] For example, the start-stop protection exit condition can be associated with upstream false detection and frame loss. As described above in the start-stop protection conditions, false detection and frame loss are directly related to the vehicle's ability to operate normally. Therefore, this can be used as a factor associated with the start-stop protection exit condition, and the false detection and frame loss of the vehicle can be quantified using the false detection rate and the number of consecutive frame losses, respectively.

[0059] For example, the start-stop protection exit condition can be linked to the real-time accuracy of ego-vehicle identification. Similar to upstream false detection and frame loss, the real-time accuracy of ego-vehicle identification is directly related to the ability of the autonomous vehicle to operate normally. Therefore, the real-time accuracy of ego-vehicle identification can also be a factor associated with the start-stop protection exit condition, and this factor can be quantified by configuring a preset error threshold.

[0060] It is worth noting that the above factors can be used alone or in combination when used to construct the start-stop protection exit conditions, and the embodiments of the present disclosure do not limit this.

[0061] In combination with the above examples, it can be seen that the above S201 that satisfies the start-stop protection exit condition can be specifically described as, in the process of executing the emergency stop operation, detecting whether the unmanned vehicle meets one or more of the following conditions: the current trajectory of the unmanned vehicle is the starting trajectory; the number of times the unmanned vehicle continuously outputs the starting trajectory reaches a preset number of times; the cumulative parking time of the unmanned vehicle reaches a preset time; the unmanned vehicle does not detect the preset parking reason; the behavior of the external entity does not interfere with the driving trajectory of the unmanned vehicle; the upstream false detection rate of the unmanned vehicle is less than the preset false detection rate threshold; the number of consecutive frame losses of the unmanned vehicle is less than the preset frame number threshold; the self-vehicle recognition error of the unmanned vehicle is less than the preset error threshold.

[0062] For example, in practical applications, the vehicle's current trajectory can be selected as the starting trajectory, and the vehicle has continuously output the starting trajectory a preset number of times as the vehicle's start-stop protection exit conditions. That is, when both conditions are met, the vehicle exits the start-stop protection mode. The following describes the principles and implementation of the start-stop protection exit conditions in this example, using a specific embodiment.

[0063] In some embodiments, S201 specifically includes: obtaining trajectory parameters of the unmanned vehicle during the emergency stop operation. Based on the trajectory parameters, determining whether the unmanned vehicle meets the start-stop protection exit conditions. The trajectory parameters include: the unmanned vehicle's current trajectory attributes and the number of times the unmanned vehicle has continuously output a starting trajectory.

[0064] Accordingly, the above S202 specifically includes: if the current trajectory of the unmanned vehicle is a starting trajectory and the number of times the unmanned vehicle continuously outputs the starting trajectory reaches a preset number of times, controlling the unmanned vehicle to exit the start-stop protection mode. The starting trajectory is a driving trajectory in which the speed parameter of at least some path points is greater than zero.

[0065] For example, the process of executing an emergency stop can be understood as a continuous process, triggered by the unmanned vehicle entering the start-stop protection mode and the execution module receiving an emergency stop command. While the unmanned vehicle is in start-stop protection mode, the decision-making system issues an emergency stop command to the execution module after each decision and output of the driving trajectory, thereby maintaining the unmanned vehicle in a stopped state. The process of the unmanned vehicle maintaining a stopped state via the emergency stop command can be understood as the process of executing an emergency stop.

[0066] For example, when acquiring the trajectory parameters of the unmanned vehicle, the trajectory parameters can be acquired periodically according to a preset frame interval. It is understood that before deciding and outputting a driving trajectory, the unmanned vehicle collects data frames (e.g., point cloud data frames, image data frames) at a certain frequency based on its onboard sensors. These data frames enable the unmanned vehicle to perceive its surrounding environmental conditions and thus decide and output a corresponding driving trajectory. The sensor acquisition interval is the preset frame interval.

[0067] When executing the start-stop control method provided by the present invention, when the unmanned vehicle is not in the start-stop protection mode, the decision system can also obtain relevant parameters for determining whether the unmanned vehicle meets the start-stop protection conditions according to a preset frame interval after outputting the driving trajectory each time, and output the judgment result.

[0068] When the unmanned vehicle is already in the start-stop protection mode, after the decision system decides to output the driving trajectory, it can also obtain relevant parameters (such as the trajectory parameters in this example) used to determine whether the unmanned vehicle meets the start-stop protection exit conditions according to the preset frame interval, and output the judgment result.

[0069] During the emergency stop operation, the autonomous vehicle can determine whether it meets the start-stop protection exit conditions based on each acquired trajectory parameter after completing the trajectory parameter acquisition. If the autonomous vehicle does not meet the start-stop protection exit conditions, it will continue to issue an emergency stop command to keep the autonomous vehicle in the start-stop protection mode.

[0070] It should be understood that this embodiment uses the unmanned vehicle's current trajectory attribute as a starting trajectory, and the number of times the unmanned vehicle continuously outputs the starting trajectory reaching a preset number of times, as the exit conditions for the start-stop protection. The current trajectory attribute being a starting trajectory means that the unmanned vehicle currently meets the starting conditions, that is, no internal or external factors that affect the unmanned vehicle's start have been detected. Internal factors can include sensor failures and frame loss, or failures in the unmanned vehicle's decision-making system, while external factors can include the presence of unavoidable obstacles in the unmanned vehicle's route. From this perspective, the unmanned vehicle's current trajectory attributes can, to a certain extent, cover the various relevant factors mentioned above.

[0071] However, because the unmanned vehicle's environmental conditions and its own state are unpredictable, significant uncertainty exists. Therefore, even if the unmanned vehicle's current trajectory output is attributed to a starting trajectory, it does not necessarily mean that the next decision output will be the same trajectory. To minimize this uncertainty and prevent the unmanned vehicle from stopping again within a short period of time after starting, this embodiment also uses the number of times the unmanned vehicle continuously outputs the starting trajectory reaching a preset number as another start-stop protection exit condition. When the unmanned vehicle continuously outputs the starting trajectory, it can be determined that the unmanned vehicle has met the starting conditions for a continuous period of time, indirectly reflecting that the internal and external interference factors affecting the unmanned vehicle's start have been steadily eliminated, and the unmanned vehicle has a high probability of normal operation after starting. Based on this, the unmanned vehicle's current trajectory attribute is a starting trajectory. This judgment condition ensures that the period during which the unmanned vehicle continuously outputs the starting trajectory is from a certain historical moment to the current moment, that is, the starting conditions are continuously met within the time range closest to the current moment.

[0072] As can be seen, the unmanned vehicle's current trajectory attribute being a starting trajectory and the number of times the unmanned vehicle has continuously output a starting trajectory reaching a preset number of times complement each other and together constitute the start-stop protection exit condition in this embodiment. Only when the unmanned vehicle meets these two conditions simultaneously is it determined that the unmanned vehicle can start and the unmanned vehicle is controlled to exit the start-stop protection mode. This configuration can not only streamline the judgment conditions to a certain extent and improve the execution efficiency of the judgment process, but also minimize the unmanned vehicle from repeatedly starting and stopping in a short period of time, providing effective protection against frequent starts and stops of the unmanned vehicle.

[0073] For example, when judging the exit condition of the start-stop protection, if the current trajectory of the unmanned vehicle is a starting trajectory and the number of times the unmanned vehicle continuously outputs the starting trajectory does not reach a preset number, the emergency stop instruction can continue to be issued; and / or, if the current trajectory of the unmanned vehicle is not a starting trajectory, the emergency stop instruction can continue to be issued.

[0074] That is to say, when any of the above two start-stop protection exit conditions is not met, the unmanned vehicle maintains the start-stop protection mode and continues to perform the emergency stop operation to maintain the parking state.

[0075] The following describes in detail how to determine the timing of exiting the start-stop protection mode in the decision-making system of the unmanned vehicle, combining the above two start-stop protection exit conditions.

[0076] Since the above two start-stop protection exit conditions are both related to the trajectory attributes of the unmanned vehicle's decision output, and the condition of "the number of times the unmanned vehicle continuously outputs the starting trajectory" focuses on both the continuity of the trajectory attributes and the number of specific trajectory attributes, the start-stop protection exit conditions can be judged by setting a counter.

[0077] In some embodiments, after the unmanned vehicle enters the start-stop protection mode, a counter can be initialized, which is used to record the number of times the unmanned vehicle continuously outputs the starting trajectory.

[0078] Next, the following steps can be repeated until the count in the counter reaches the preset threshold, and the unmanned vehicle is controlled to exit the start-stop protection mode: When the trajectory output by the unmanned vehicle is a starting trajectory, the count in the counter can be increased by one.

[0079] When the trajectory output by the unmanned vehicle is a non-starting trajectory, the count in the counter can be cleared.

[0080] For example, each time the counter is incremented by one, an emergency stop instruction may be issued to prevent the unmanned vehicle from starting due to the output starting trajectory.

[0081] For example, an emergency stop command can be issued each time the counter is cleared. Although non-starting trajectories do not directly cause the autonomous vehicle to start, issuing a higher-priority emergency stop command can still prevent the autonomous vehicle from starting due to other reasons. These other reasons may include start commands from the upstream dispatch system.

[0082] Thus, the solution provided by the embodiment of the present disclosure can put the unmanned vehicle into the start-stop protection mode when it is in a parked state and meets the start-stop protection conditions, thereby preventing the unmanned vehicle from frequently starting and stopping in a short period of time. While the unmanned vehicle is in the start-stop protection mode, the decision-making system of the unmanned vehicle can issue an emergency stop command to the execution module to ensure that the vehicle is stationary, and this does not conflict with the normal decision-making output of the driving trajectory by the decision-making system. When the decision-making system continuously outputs the starting trajectory for a certain period of time adjacent to the current moment, the unmanned vehicle can exit the start-stop protection mode and start the starting behavior according to the output starting trajectory.

[0083] Based on the same inventive concept, the present disclosure also provides a start-stop control device for an unmanned vehicle, such as the following embodiment. Figure 1 The method embodiment shown is similar, so the implementation of the unmanned vehicle start-stop control device embodiment can refer to the above Figure 1 The implementation of the illustrated method embodiment will not be repeated any more.

[0084] Figure 3 FIG. 1 is a schematic diagram showing the structure of a start-stop control device for an unmanned vehicle according to an embodiment of the present disclosure. Figure 3 As shown, the unmanned vehicle start-stop control device 300 includes: Detection module 301 is used to detect whether the unmanned vehicle meets the start-stop protection conditions when the unmanned vehicle enters the parking state. The start-stop protection conditions are related to one or more of the following factors: the reason for the unmanned vehicle to stop; the current speed of the unmanned vehicle; the current trajectory attributes of the unmanned vehicle; the behavior characteristics of external entities; upstream false detection and frame loss; and the real-time accuracy of self-vehicle identification.

[0085] The control module 302 is used to control the unmanned vehicle to enter the start-stop protection mode and issue an emergency stop instruction when the unmanned vehicle meets the start-stop protection conditions. The emergency stop instruction is used to instruct the unmanned vehicle to perform an emergency stop operation.

[0086] In some embodiments, the detection module 301 is further configured to detect whether the unmanned vehicle meets the start-stop protection exit conditions during an emergency stop operation. The start-stop protection exit conditions may be related to one or more of the following factors: the unmanned vehicle's current trajectory attributes; the number of times the unmanned vehicle continuously outputs a starting trajectory; the unmanned vehicle's cumulative parking time; the reason for the unmanned vehicle's parking; the behavioral characteristics of external entities; upstream false detection and frame loss; and the real-time accuracy of self-vehicle identification. The control module 302 is further configured to control the unmanned vehicle to exit the start-stop protection mode if the unmanned vehicle meets the start-stop protection exit conditions.

[0087] In some embodiments, the detection module 301 is further configured to obtain the unmanned vehicle's trajectory parameters during the emergency stop operation. The trajectory parameters include: the unmanned vehicle's current trajectory attributes and the number of times the unmanned vehicle has continuously output a starting trajectory; based on the trajectory parameters, the detection module 301 is configured to determine whether the unmanned vehicle meets the start-stop protection exit conditions. The control module 302 is further configured to control the unmanned vehicle to exit the start-stop protection mode if the unmanned vehicle's current trajectory is a starting trajectory and the number of times the unmanned vehicle has continuously output a starting trajectory reaches a preset number; the starting trajectory is a driving trajectory in which the speed parameter of at least some path points is greater than zero.

[0088] In some embodiments, the control module 302 is further configured to, in a case where the current trajectory of the unmanned vehicle is a start-up trajectory and the number of times that the start-up trajectory is continuously output by the unmanned vehicle does not reach a preset number of times, continue to issue the emergency stop instruction; and / or in a case where the current trajectory of the unmanned vehicle is a non-start-up trajectory, continue to issue the emergency stop instruction.

[0089] In some embodiments, the control module 302 is further configured to initialize a counter, the counter being configured to record the number of times that the start-up trajectory is continuously output by the unmanned vehicle; repeatedly perform the following steps until the count in the counter is equal to a preset threshold, and control the unmanned vehicle to exit the start-stop protection mode: in a case where the trajectory output by the unmanned vehicle is a start-up trajectory, increase the count in the counter by one; in a case where the trajectory output by the unmanned vehicle is a non-start-up trajectory, clear the count in the counter.

[0090] In some embodiments, the detection module 301 is further configured to periodically acquire the trajectory parameters of the unmanned vehicle at a preset frame interval; determine whether the unmanned vehicle satisfies the start-stop protection exit condition based on the trajectory parameters, including: after each acquisition of the trajectory parameters, determining whether the unmanned vehicle satisfies the start-stop protection exit condition based on the acquired trajectory parameters; and in a case where the unmanned vehicle does not satisfy the start-stop protection exit condition, continuing to issue the emergency stop instruction.

[0091] In some embodiments, the detection module 301 is further configured to, during the execution of the emergency stop operation, detect whether the unmanned vehicle satisfies one or more of the following conditions: the current trajectory of the unmanned vehicle is a start-up trajectory; the number of times that the start-up trajectory is continuously output by the unmanned vehicle reaches a preset number of times; the cumulative stop time length of the unmanned vehicle reaches a preset time length; the unmanned vehicle does not detect a preset stop reason; the behavior of an external entity does not interfere with the driving trajectory of the unmanned vehicle; the upstream false detection rate of the unmanned vehicle is less than a preset false detection rate threshold; the number of consecutive frame losses of the unmanned vehicle is less than a preset frame number threshold; and the self-vehicle identification error of the unmanned vehicle is less than a preset error threshold.

[0092] In some embodiments, the detection module 301 is further configured to, in a case where the unmanned vehicle enters a stop state, detect whether the unmanned vehicle satisfies one or more of the following conditions: the unmanned vehicle enters the stop state due to a preset stop reason; the current speed of the unmanned vehicle is less than a preset speed threshold; the current trajectory of the unmanned vehicle is a non-start-up trajectory; the behavior of an external entity interferes with the driving trajectory of the unmanned vehicle; the upstream false detection rate of the unmanned vehicle is greater than a preset false detection rate threshold; the number of consecutive frame losses of the unmanned vehicle is greater than a preset frame number threshold; and the self-vehicle identification error of the unmanned vehicle is greater than a preset error threshold.

[0093] Those skilled in the art will appreciate that various aspects of the present disclosure may be implemented as systems, methods, or program products. Therefore, various aspects of the present disclosure may be implemented in the following forms: entirely in hardware, entirely in software (including firmware, microcode, etc.), or in a combination of hardware and software, collectively referred to herein as "circuits," "modules," or "systems."

[0094] Based on the same inventive concept, the present disclosure also provides an unmanned vehicle for executing the unmanned vehicle start-stop control method described in the above embodiment. Figure 1 The implementation of the illustrated method embodiment will not be repeated any more.

[0095] Through the description of each embodiment above, it will be readily understood by those skilled in the art that the example embodiments described herein can be implemented via software or via a combination of software and necessary hardware. Therefore, the technical solutions according to the embodiments of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, or mobile hard drive) or on a network and includes several instructions for enabling a computing device (such as a personal computer, server, mobile terminal, or network device) to execute the methods according to the embodiments of the present disclosure.

[0096] Other embodiments of the present disclosure will readily occur to those skilled in the art after considering the specification and practicing the invention disclosed herein. This disclosure is intended to cover any variations, uses, or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary techniques in the art not disclosed herein. The description and examples are to be considered as exemplary only, with the true scope and spirit of the present disclosure being indicated by the appended claims.

Claims

1. A method for controlling the start and stop of an unmanned vehicle, characterized in that: include: When the unmanned vehicle enters a parking state, detecting whether the unmanned vehicle meets the start-stop protection conditions; wherein the start-stop protection conditions are related to one or more of the following factors: the reason for the unmanned vehicle to stop; the current speed of the unmanned vehicle; the current trajectory attributes of the unmanned vehicle; the behavior characteristics of external entities; the upstream false detection frame loss situation; the real-time accuracy of the self-vehicle identification; When the unmanned vehicle meets the start-stop protection condition, the unmanned vehicle is controlled to enter the start-stop protection mode and an emergency stop instruction is issued; wherein, the emergency stop instruction is used to instruct the unmanned vehicle to perform an emergency stop operation.

2. The method according to claim 1, characterized in that The method further comprises: During the emergency stop operation, the unmanned vehicle is detected to determine whether it meets the start-stop protection exit condition; wherein the start-stop protection exit condition is related to one or more of the following factors: the current trajectory attributes of the unmanned vehicle; the number of times the unmanned vehicle continuously outputs the starting trajectory; the accumulated parking time of the unmanned vehicle; the reason for the unmanned vehicle's parking; the behavioral characteristics of external entities; the upstream false detection frame loss situation; and the real-time accuracy of self-vehicle identification. When the unmanned vehicle meets the start-stop protection exit condition, the unmanned vehicle is controlled to exit the start-stop protection mode.

3. The method according to claim 2, characterized in that During the emergency stop operation, detecting whether the unmanned vehicle meets the start-stop protection exit condition includes: During the emergency stop operation, obtaining trajectory parameters of the unmanned vehicle, the trajectory parameters including: current trajectory attributes of the unmanned vehicle and the number of times the unmanned vehicle continuously outputs a starting trajectory; Based on the trajectory parameters, determining whether the unmanned vehicle meets the start-stop protection exit condition; When the unmanned vehicle satisfies the start-stop protection exit condition, controlling the unmanned vehicle to exit the start-stop protection mode includes: When the current trajectory of the unmanned vehicle is a starting trajectory and the number of times the unmanned vehicle continuously outputs the starting trajectory reaches a preset number, controlling the unmanned vehicle to exit the start-stop protection mode; The starting trajectory is a driving trajectory in which the speed parameter of at least some path points is greater than zero.

4. The method according to claim 3, wherein The method further comprises: If the current trajectory of the unmanned vehicle is a starting trajectory and the number of times the unmanned vehicle continuously outputs the starting trajectory does not reach a preset number, continue to issue an emergency stop instruction; and / or When the current trajectory of the unmanned vehicle is not a starting trajectory, the emergency stop instruction continues to be issued.

5. The method according to claim 3, characterized in that When the current trajectory of the unmanned vehicle is a starting trajectory and the number of times the unmanned vehicle continuously outputs the starting trajectory reaches a preset number, controlling the unmanned vehicle to exit the start-stop protection mode includes: Initializing a counter, the counter being used to record the number of times the unmanned vehicle continuously outputs the starting trajectory; Repeat the following steps until the count in the counter is equal to a preset threshold, and then control the unmanned vehicle to exit the start-stop protection mode: When the trajectory output by the unmanned vehicle is a starting trajectory, the count in the counter is increased by one; When the trajectory output by the unmanned vehicle is a non-starting trajectory, the count in the counter is cleared.

6. The method according to claim 3, characterized in that The obtaining of the trajectory parameters of the unmanned vehicle includes: Periodically acquiring trajectory parameters of the unmanned vehicle according to a preset frame interval; The determining, based on the trajectory parameters, whether the unmanned vehicle satisfies a start-stop protection exit condition includes: After each completion of trajectory parameter acquisition, determining whether the unmanned vehicle meets the start-stop protection exit condition based on the acquired trajectory parameters; When the unmanned vehicle does not meet the start-stop protection exit condition, the emergency stop instruction continues to be issued.

7. The method according to claim 2, characterized in that During the emergency stop operation, detecting whether the unmanned vehicle meets the start-stop protection exit condition includes: During the emergency stop operation, the unmanned vehicle is detected to determine whether it meets one or more of the following conditions: The current trajectory of the unmanned vehicle is the starting trajectory; The number of times the unmanned vehicle continuously outputs the starting trajectory reaches a preset number; The accumulated parking time of the unmanned vehicle reaches a preset time; The unmanned vehicle fails to detect a preset parking reason; The actions of external entities do not interfere with the driving trajectory of the autonomous vehicle; The upstream false detection rate of the unmanned vehicle is less than a preset false detection rate threshold; The number of consecutive frame losses of the unmanned vehicle is less than a preset frame threshold; The self-vehicle recognition error of the unmanned vehicle is less than a preset error threshold.

8. The method according to claim 1, characterized in that When the unmanned vehicle enters a parking state, detecting whether the unmanned vehicle meets the start-stop protection condition includes: When the unmanned vehicle enters the parking state, it is detected whether the unmanned vehicle meets one or more of the following conditions: The unmanned vehicle enters a parking state due to a preset parking reason; The current speed of the unmanned vehicle is less than a preset speed threshold; The current trajectory of the unmanned vehicle is a non-starting trajectory; The behavior of an external entity interferes with the driving trajectory of the autonomous vehicle; The upstream false detection rate of the unmanned vehicle is greater than a preset false detection rate threshold; The number of consecutive frame drops of the unmanned vehicle is greater than a preset frame threshold; The self-vehicle recognition error of the unmanned vehicle is greater than a preset error threshold.

9. A start-stop control device for an unmanned vehicle, characterized in that: include: A detection module is configured to detect whether the unmanned vehicle meets the start-stop protection conditions when the unmanned vehicle enters a parking state; wherein the start-stop protection conditions are related to one or more of the following factors: the reason for the unmanned vehicle to stop; the current speed of the unmanned vehicle; the current trajectory attributes of the unmanned vehicle; the behavioral characteristics of external entities; the upstream false detection frame loss situation; and the real-time accuracy of self-vehicle identification. A control module is used to control the unmanned vehicle to enter a start-stop protection mode and issue an emergency stop instruction when the unmanned vehicle meets the start-stop protection conditions; wherein the emergency stop instruction is used to instruct the unmanned vehicle to perform an emergency stop operation.

10. An unmanned vehicle, characterized in that: The unmanned vehicle is used to execute the method according to any one of claims 1 to 8.

Images