Parking control method and device, electronic equipment and medium

Abstract

The application provides a parking control method and device, electronic equipment and medium, and relates to the technical field of vehicles. The method comprises the following steps: acquiring a parking drivable distance of a vehicle according to a current motion state of the vehicle and sensing information of a parking obstacle; acquiring a parking reference trajectory according to the parking drivable distance and a current desired speed sequence of the vehicle in the case that it is determined that there is no collision risk of the vehicle according to the parking drivable distance, wherein the desired speed sequence comprises vehicle desired speeds of the vehicle at a series of future time points, and the parking reference trajectory comprises vehicle longitudinal reference positions and vehicle reference speeds at the series of future time points; acquiring a first acceleration for controlling the vehicle to park by using a prediction function control algorithm according to the current motion state of the vehicle and the parking reference trajectory; and controlling the vehicle to park according to the first acceleration. The application is helpful to realize stable control of the vehicle speed in an automatic parking scene.

Description

Technical Field

[0001] This application relates to the field of vehicle technology, and in particular to a parking control method, device, electronic device, and medium. Background Technology

[0002] Automated parking is a driver assistance technology that enables cars to park themselves. During automated parking, the accelerator, brakes, and steering are all taken over by the vehicle's automated driving system. Using automated parking technology can help drivers park their vehicles more easily and safely in narrow or complex parking spaces.

[0003] Automatic parking systems typically operate at extremely low speeds (e.g., <1m / s), and vehicles are prone to errors and fluctuations when operating at extremely low speeds. Therefore, the requirements for speed control in automatic parking scenarios are extremely high. Summary of the Invention

[0004] This application provides a parking control method, device, electronic device, and medium, which helps to achieve stable control of vehicle speed in automatic parking scenarios.

[0005] In a first aspect, embodiments of this application provide a parking control method, comprising: obtaining a drivable parking distance of the vehicle based on the vehicle's current motion state and perception information of parking obstacles; if it is determined that the vehicle has no collision risk based on the drivable parking distance, obtaining a parking reference trajectory based on the drivable parking distance and the vehicle's current expected speed sequence, wherein the expected speed sequence includes the vehicle's expected speed at a series of future time points, and the parking reference trajectory includes the vehicle's longitudinal reference position and reference speed at a series of future time points; using a predictive function control algorithm, obtaining a first acceleration for controlling the vehicle's parking based on the vehicle's current motion state and the parking reference trajectory; and controlling the vehicle's parking based on the first acceleration.

[0006] This application applies the predictive function control algorithm to the vehicle parking control scenario. The predictive function control algorithm can make the vehicle parking acceleration more regular. Thus, by using the acceleration output by the predictive function control algorithm to control the vehicle parking in the absence of collision risk, stable control of vehicle speed can be achieved in the automatic parking scenario, which helps to achieve the stability and safety of vehicle parking.

[0007] Optionally, a predictive function control algorithm is used to obtain a first acceleration for controlling vehicle parking based on the vehicle's current motion state and parking reference trajectory. This includes: inputting the vehicle's current motion state into the predictive model to obtain the first acceleration output by the predictive model under the constraints of basis functions and optimization objectives; wherein, the optimization objectives include ensuring that the error between the predicted parking trajectory and the parking reference trajectory under the first acceleration is not greater than the error between the predicted parking trajectory and the parking reference trajectory under other accelerations; the predicted parking trajectory includes the predicted longitudinal position and predicted speed of the vehicle at a series of future time points; the basis functions are used to ensure that the acceleration remains consistent at a series of future time points, and the predictive model is a second-order vehicle model regarding the vehicle's longitudinal position and speed.

[0008] Based on the characteristics of predictive function control algorithms in dealing with errors and fluctuations, this application embodiment uses predictive function control algorithms in the absence of parking collision risk, and combines the vehicle's current actual motion state and parking reference trajectory to achieve parking control, which helps to achieve stable control of vehicle speed in automatic parking scenarios.

[0009] Optionally, based on the vehicle's current motion state and the perception information of parking obstacles, the vehicle's drivable parking distance is obtained, including: obtaining the vehicle's current turning radius based on the vehicle's current motion state; obtaining a first parking trajectory line based on the turning radius, wherein the radius of curvature of the first parking trajectory line is the turning radius, the starting point of the first parking trajectory line is the vehicle's current position, and the longitudinal position of the ending point of the first parking trajectory line is the longitudinal position of the parking end position of the desired parking trajectory line planned by the vehicle's planning module; obtaining a second parking trajectory line based on the turning radius and the perception information of parking obstacles, wherein the radius of curvature of the second parking trajectory line is the turning radius, the starting point of the second parking trajectory line is the vehicle's current position, and if the perception information indicates that the vehicle perceives a parking obstacle, then the ending point of the second parking trajectory line is the intersection of the perceived parking obstacle and the second parking trajectory line; obtaining a length set, the length set including the length of the first parking trajectory line and the length of the second parking trajectory line; and determining the minimum length in the length set as the vehicle's drivable parking distance.

[0010] The first parking trajectory line reflects the longitudinal distance traveled from the vehicle's current position to the end position of the parking maneuver (i.e., the remaining distance to the endpoint) without considering parking obstacles. The second parking trajectory line reflects the longitudinal distance traveled from the vehicle's current position to the location of the parking obstacle (i.e., the remaining distance to the obstacle) when considering parking obstacles. By comparing the magnitudes of the remaining distance to the endpoint and the remaining distance to the obstacle, the feasible parking distance can be accurately determined, thereby supporting the accurate assessment of collision risk.

[0011] Optionally, the second parking trajectory line includes parking trajectory lines for multiple corner points of the vehicle. This helps prevent the vehicle's edges from scraping against obstacles during parking.

[0012] Optionally, a parking reference trajectory is obtained based on the parking drivable distance and the vehicle's current expected speed sequence, including: if the second parking trajectory line has a minimum length, obtaining the vehicle's longitudinal reference position based on the vehicle's longitudinal position in the second parking trajectory line.

[0013] Since the second parking trajectory line has a minimum length, it can be assumed that there are parking obstacles that affect the vehicle's parking but do not pose a collision risk at the moment. If it is expected that the parking trajectory line does not involve avoiding the parking obstacle in the longitudinal direction, then the longitudinal reference position of the vehicle can be obtained based on the longitudinal position of the vehicle in the second parking trajectory line, so as to ensure that the vehicle can continue to park for a period of time afterward without colliding with the parking obstacle.

[0014] Optionally, if the perception information indicates that the vehicle has not perceived a parking obstacle, the length of the second parking trajectory line is greater than the length of the first parking trajectory line; the parking reference trajectory is obtained based on the parking drivable distance and the vehicle's current expected speed sequence, including: if the first parking trajectory line has a minimum length, obtaining the vehicle's longitudinal reference position based on the vehicle's longitudinal position in the first parking trajectory line.

[0015] Since the first parking trajectory line has the minimum length, it can be assumed that there are no parking obstacles that affect the parking of the vehicle. Therefore, the longitudinal reference position of the vehicle can be obtained based on the longitudinal position of the vehicle in the first parking trajectory line, so as to ensure that the vehicle can continue to park for a period of time and usually will not collide with obstacles.

[0016] Optionally, the length set also includes: a drivable parking length, which is planned by the vehicle's planning module based on the perception information of parking obstacles; and obtaining a parking reference trajectory based on the drivable parking distance and the vehicle's current expected speed sequence, including: if the drivable parking length is the minimum length, obtaining the vehicle's longitudinal reference position based on the expected speed sequence output by the vehicle's planning module and the vehicle's current motion state.

[0017] Since the drivable length for parking is the minimum length, it can be assumed that the planning module senses parking obstacles that affect the vehicle's parking but do not pose a collision risk at the moment. If the desired parking trajectory line does not involve avoiding the parking obstacle in the longitudinal direction, the longitudinal reference position of the vehicle can be obtained based on the vehicle's current motion state (such as the vehicle's current speed and acceleration) and the planned desired speed sequence, so as to ensure that the vehicle can continue to park for a period of time afterward without colliding with the parking obstacle.

[0018] Optionally, the parking control method further includes: inputting the acquired drivable parking distance and the vehicle's current parking-related information into a learned neural network to obtain the output of the learned neural network, the output of which is used to describe whether the vehicle has a collision risk; wherein, the learned neural network is obtained by performing normalized exponential function regression learning using a neural network, and the data used for regression learning includes the drivable parking distance and parking-related information of the test vehicle in a parking scenario with a collision risk, and the drivable parking distance and parking-related information of the test vehicle in a parking scenario without a collision risk; the parking-related information includes at least one of the following: road surface slope information of the road where the vehicle is located, the vehicle's motion state, and the area information of the vehicle's parking area.

[0019] By performing neural network regression learning based on test data under different parking scenarios, and inputting the currently acquired drivable parking distance and parking-related information into the learned neural network, it is possible to accurately determine whether a vehicle is at risk of a parking collision.

[0020] Optionally, the parking control method also includes: applying the brakes to the vehicle when a collision risk is determined based on the available parking distance.

[0021] During vehicle parking, if a collision risk is assessed based on the available parking distance, parking safety will be prioritized, and the vehicle will be braked to ensure parking safety. If there is no collision risk, parking stability will be prioritized, thus comprehensively ensuring both safety and stability requirements during the parking process.

[0022] Secondly, embodiments of this application provide a parking control device, comprising: a first acquisition module, configured to acquire the vehicle's drivable parking distance based on the vehicle's current motion state and perception information of parking obstacles; a second acquisition module, configured to acquire a parking reference trajectory based on the drivable parking distance and the vehicle's current expected speed sequence, wherein the expected speed sequence includes the vehicle's expected speed at a series of future time points, and the parking reference trajectory includes the vehicle's longitudinal reference position and reference speed at a series of future time points; a third acquisition module, configured to acquire a first acceleration for controlling the vehicle's parking based on the vehicle's current motion state and parking reference trajectory using a predictive function control algorithm; and a control module, configured to control the vehicle's parking based on the first acceleration.

[0023] Thirdly, embodiments of this application provide an electronic chip, including: a processor for executing computer program instructions stored in a memory, wherein when the computer program instructions are executed by the processor, the electronic chip is triggered to perform the method as described in any of the first aspects.

[0024] Fourthly, embodiments of this application provide an electronic device including at least one processor and a memory coupled together. The memory is used to store computer program instructions, and the processor is used to execute the computer program instructions. When the computer program instructions are executed by the processor, the electronic device is triggered to perform a method as described in any of the first aspects.

[0025] Fifthly, embodiments of this application provide a computer-readable storage medium storing a computer program that, when run on a computer, causes the computer to perform the method as described in any of the first aspects.

[0026] In a sixth aspect, embodiments of this application provide a computer program product, which includes a computer program that, when run on a computer, causes the computer to perform the method as described in any of the first aspects.

[0027] The technical effects of the aforementioned aspects can be referenced from each other, and will not be elaborated further here. Attached Figure Description

[0028] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly described below.

[0029] Figure 1 A schematic flowchart illustrating a parking control method provided in an embodiment of this application;

[0030] Figure 2 A schematic diagram illustrating the acquisition of the parking distance in a given scenario, as provided in an embodiment of this application.

[0031] Figure 3 This is a schematic diagram illustrating another scenario for obtaining the parking distance in an embodiment of this application.

[0032] Figure 4 A block diagram of a parking control device provided in an embodiment of this application;

[0033] Figure 5 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0034] To better understand the technical solution of this application, the embodiments of this application will be described in detail below with reference to the accompanying drawings.

[0035] It should be understood that the described embodiments are merely some, not all, of the embodiments in this application. All other embodiments obtained by those skilled in the art based on the embodiments in this application without inventive effort are within the scope of protection of this application.

[0036] The terminology used in the embodiments of this application is for the purpose of describing particular embodiments only and is not intended to be limiting of this application. The singular forms “a,” “the,” and “the” used in the embodiments of this application and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise.

[0037] It should be understood that the term "at least one" as used in this document refers to one or more, and "more than one" refers to two or more. The term "and / or" as used in this document is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, and B alone. A and B can be singular or plural. Additionally, the character " / " in this document generally indicates that the preceding and following related objects are in an "or" relationship. "At least one of the following" and similar expressions refer to any combination of these items, including any combination of singular or plural items. For example, at least one of a, b, and c can represent: a, b, c, ab, ac, bc, or abc, where a, b, and c can be single or multiple.

[0038] It should be understood that although the terms "first," "second," etc., may be used to describe the set thresholds in the embodiments of this application, these set thresholds should not be limited to these terms. These terms are only used to distinguish the set thresholds from each other. For example, without departing from the scope of the embodiments of this application, the first set threshold may also be referred to as the second set threshold, and similarly, the second set threshold may also be referred to as the first set threshold.

[0039] like Figure 1 As shown, this application provides a parking control method, which may include steps 101 to 104. In one embodiment, the entity executing the parking control method can be a vehicle with autonomous driving capabilities, and the vehicle can implement parking control based on its autonomous driving system. In one embodiment, the vehicle can be a four-wheeled car.

[0040] In one embodiment, a parking control method can be periodically executed during vehicle parking to achieve stable control of vehicle speed in an automatic parking scenario.

[0041] Step 101: Based on the vehicle's current motion state and the perception information of parking obstacles, obtain the vehicle's parking distance.

[0042] Feasibly, the vehicle may include a planning module. This module, when the vehicle needs to park, can plan the desired parking trajectory by combining the vehicle's motion state (such as vehicle position, speed, and heading angle) and its environment (such as whether there are obstacles in front of the vehicle). For example, in Figure 2 In the parking scenario shown without parking obstacles, the desired parking trajectory planned by the vehicle's planning module can be as follows: Figure 2 The curve shown in reference number 202.

[0043] Normally, parking can be achieved by driving along the desired parking trajectory. However, due to factors such as sensor perception results (e.g., failure to detect some obstacles), the planning strategy of the planning module (e.g., ignoring or weakening the impact of some obstacles on the parking trajectory), and the possibility of changing obstacle positions (e.g., pedestrian movement), there is still a risk of collision while driving along the desired parking trajectory. For example, in... Figure 3 In the parking scenario shown with parking obstacle 304, the desired parking trajectory planned by the vehicle's planning module can be as follows: Figure 3 The curve shown in reference number 302.

[0044] To avoid parking collisions, during the parking process, the vehicle's current driving distance can be periodically determined based on its current movement and perception of parking obstacles. This driving distance can then be used to assess the potential collision risk. If there is no collision risk, parking continues; if there is a collision risk, the vehicle's brakes are applied.

[0045] During vehicle parking, if a collision risk is assessed based on the available parking distance, parking safety is prioritized, and the vehicle is braked to ensure parking safety. Conversely, if there is no collision risk, parking stability is prioritized, thus comprehensively ensuring both safety and stability requirements during parking. Therefore, in one embodiment of this application, the parking control method further includes: applying braking to the vehicle when a collision risk is determined based on the available parking distance. In one feasible implementation, this braking action can be a complete stop.

[0046] If braking is applied to the vehicle in the event of a collision risk, and continuing to park would result in a collision or scrape, one feasible approach is to request the planning module to replan the desired parking trajectory after braking is applied in the event of a collision risk. Subsequent parking control is then implemented based on the newly planned desired parking trajectory.

[0047] In one embodiment, the vehicle's motion status can be collected in real time using sensors (such as speed sensors, positioning sensors, etc.) used to sense the vehicle's state.

[0048] In one embodiment, the current motion state of the vehicle may include some or all of the actual motion information of the vehicle, such as wheel angle, vehicle speed, vehicle acceleration, vehicle position (e.g., the vehicle's lateral and longitudinal coordinates), and vehicle heading angle.

[0049] In one embodiment, parking obstacles (such as obstacles located in the parking path) that affect the vehicle's parking can be detected in real time using sensors (such as radar, cameras, etc.) used by the vehicle to perceive its environment. Parking obstacles can be, for example, pedestrians, objects, etc.

[0050] If there are no parking obstacles in the environment, the vehicle's perception information about parking obstacles can indicate that the vehicle is not currently sensing any parking obstacles. If there are parking obstacles in the environment, and the sensors are able to detect them, the vehicle's perception information about parking obstacles can indicate that the vehicle is currently sensing any parking obstacles, and may include information about the sensed parking obstacles (such as their location).

[0051] In one embodiment, the perception information of parking obstacles can be obtained based on the ultrasonic radar signals collected by the vehicle's onboard radar. This enables accurate perception of parking obstacles, thereby avoiding collisions between the vehicle and obstacles during parking and ensuring parking safety.

[0052] After obtaining the vehicle's current motion status and perception information of parking obstacles, the vehicle's current drivable parking distance can be determined, and the potential collision risk can be assessed accordingly. For example, if the perceived parking obstacle is close to the vehicle, the current drivable parking distance is short, and continuing to park may pose a collision risk. Conversely, if no parking obstacle is perceived or the perceived obstacle is far from the vehicle, the current drivable parking distance is long, and continuing to park usually does not pose a collision risk.

[0053] To illustrate a feasible method for obtaining the parking distance, in one embodiment of this application, step 101 may include the following steps 1011 to 1015:

[0054] Step 1011: Obtain the current turning radius of the vehicle based on its current motion state.

[0055] In one embodiment, the vehicle's motion state may include the vehicle's wheel angles, from which the vehicle's turning radius can be calculated.

[0056] Step 1012: Obtain the first parking trajectory line based on the turning radius. The radius of curvature of the first parking trajectory line is the turning radius, the starting point of the first parking trajectory line is the current position of the vehicle, and the longitudinal position of the ending point of the first parking trajectory line is the longitudinal position of the desired parking end position planned by the vehicle's planning module. The longitudinal position can be a longitudinal distance based on the vehicle coordinate system.

[0057] The starting point of the desired parking trajectory planned by the planning module can be the vehicle's position when the planning begins, and the ending point of the desired parking trajectory can be the vehicle's parking end position.

[0058] In addition to planning the desired parking trajectory, the planning module can also periodically output the planned desired speed sequence, which is the planned speed of the vehicle at various points in time within a future period (e.g., within 2 seconds).

[0059] In one feasible implementation, the first parking trajectory line is obtained by taking the vehicle's current position as the starting point, the longitudinal position of the parking end position as the ending point, and the current turning radius of the vehicle as the radius of curvature for circular expansion calculation. The first parking trajectory line is used to reflect the length of the vehicle's longitudinal driving distance when parking without considering parking obstacles; therefore, the length of the first parking trajectory line is also called the remaining distance at the end point.

[0060] exist Figure 2 In the parking scenario shown, the first parking trajectory line can be as follows: Figure 2 Curve numbered 203. In Figure 3 In the parking scenario shown, the first parking trajectory line can be as follows: Figure 3 The curve shown in reference number 303.

[0061] Please refer to Figure 2 If the parking trajectory line is expected to reflect the change in the position of the rear axle center of the vehicle, and the parking end position can be the parking end position of the rear axle center of the vehicle, then the first parking trajectory line can be a parking trajectory line that reflects the change in the position of the rear axle center of the vehicle.

[0062] Step 1013: Based on the turning radius and the perception information of parking obstacles, obtain the second parking trajectory line, wherein the radius of curvature of the second parking trajectory line is the turning radius, the starting point of the second parking trajectory line is the current position of the vehicle, and if the perception information indicates that the vehicle perceives a parking obstacle, then the ending point of the second parking trajectory line is the intersection of the perceived parking obstacle and the second parking trajectory line.

[0063] In one feasible implementation, starting from the vehicle's current position and considering the positions of existing parking obstacles, a second parking trajectory line can be obtained by expanding the circle using the vehicle's current turning radius as the radius of curvature. This second parking trajectory line reflects the length of the vehicle's longitudinal drivable distance when parking obstacles are taken into account; therefore, the length of the second parking trajectory line is also referred to as the remaining distance beyond the obstacles.

[0064] In this way, the risk of a parking collision can be reflected in the future based on the comparison between the remaining distance to the destination and the remaining distance to the obstacle, that is, the comparison between the lengths of the first parking trajectory line and the second parking trajectory line.

[0065] exist Figure 2 In the parking scenario shown, the second parking trajectory line can be as follows: Figure 2 Curve numbered 201. In Figure 3 In the parking scenario shown, the second parking trajectory line can be as follows: Figure 3 The curve indicated by reference numeral 301 intersects with obstacle 304. This means that if the vehicle parks along the desired parking trajectory, a collision will occur between the vehicle's left front corner and surrounding area and obstacle 304.

[0066] To prevent the vehicle's edge from scraping against obstacles during parking, the second parking trajectory line can be the parking trajectory line of the vehicle's corner point. Figure 2 The second parking trajectory line shown in the reference numeral 201 and Figure 3 The second parking trajectory lines shown in reference numeral 301 can all be the trajectory lines of the left front corner point of the vehicle.

[0067] To prevent the vehicle's edge from scraping against obstacles during parking, in one embodiment of this application, the second parking trajectory line includes parking trajectory lines for multiple corner points of the vehicle. These multiple corner points can be, for example, corner points located at the front left, front right, rear left, and rear right positions of the vehicle, so step 1013 can obtain four second parking trajectory lines.

[0068] In one embodiment, the lengths of the four second parking trajectory lines can be obtained separately, and the minimum length among them can be compared with the length of the first parking trajectory line. In another embodiment, the lengths of the four second parking trajectory lines can be directly compared with the length of the first parking trajectory line.

[0069] Step 1014: Obtain the length set, which includes the length of the first parking trajectory line and the length of the second parking trajectory line.

[0070] Please refer to Figure 3When a parking obstacle 304 exists, the length of the second parking trajectory line, curve 301, is less than the length of the first parking trajectory line. Therefore, in one embodiment, the current parking distance that the vehicle can travel can be the length of curve 301.

[0071] In one embodiment, please refer to Figure 2 If there are no parking obstacles that truncate curve 201, the length of curve 201 can be infinitely large, such that the length of this second parking trajectory line is greater than the length of the first parking trajectory line. In one embodiment, the current parking distance that the vehicle can travel can be the length of curve 203.

[0072] Thus, in one embodiment of this application, if the perception information indicates that the vehicle has not perceived a parking obstacle, the length of the second parking trajectory line can be greater than the length of the first parking trajectory line, thereby supporting the accurate determination of the parking distance.

[0073] Please refer to Figure 3 The first parking trajectory line reflects the longitudinal distance traveled from the vehicle's current position to the end position of the parking maneuver (i.e., the remaining distance to the end point) without considering parking obstacles. The second parking trajectory line reflects the longitudinal distance traveled from the vehicle's current position to the position of the parking obstacle (i.e., the remaining distance to the obstacle) when considering parking obstacles. By comparing the magnitudes of the remaining distance to the end point and the remaining distance to the obstacle, the feasible parking distance can be accurately determined, thereby supporting the accurate assessment of collision risk.

[0074] The presence or absence of obstacles and their impact on the available parking distance can be considered not only in the length of the second parking trajectory line but also in the parking distance planned by the vehicle's planning module. For example, when a pedestrian is in front of the vehicle while parking, the planning module can plan the parking distance based on the pedestrian's perception information, and this planned parking distance can be the longitudinal distance between the vehicle and the pedestrian. For instance, the parking distance is greater when the pedestrian is farther from the vehicle than when the pedestrian is closer to the vehicle.

[0075] For example, when there are no parking obstacles in front of the vehicle, the parking travel length planned by the planning module can be the length from the current position of the vehicle to the end position of the parking trajectory.

[0076] Thus, in one embodiment of this application, the length set further includes: the drivable parking length (or planned remaining distance), which is planned by the vehicle's planning module based on the perception information of parking obstacles. That is, the drivable parking distance can be determined by comparing the magnitudes of the remaining distance to the destination, the remaining distance to the obstacle, and the planned remaining distance, thereby supporting accurate assessment of collision risk.

[0077] In one feasible implementation, the obstacle perception information used by the planning module and the obstacle perception information used to generate the second parking trajectory line can be the same information.

[0078] In another feasible implementation, the obstacle perception information used by the planning module and the obstacle perception information used to generate the second parking trajectory line can be different information. For example, the obstacle perception information used by the planning module can be the perception image of the vehicle camera, while the obstacle perception information used to generate the second parking trajectory line can be the perception signal of the vehicle radar.

[0079] Step 1015: Determine the minimum length in the length set as the vehicle's parking distance.

[0080] In one embodiment, the length set includes the length of a first parking trajectory line and the length of a second parking trajectory line. Based on the comparison of the lengths of the first and second parking trajectory lines, the drivable distance for parking can be accurately determined, thereby supporting accurate assessment of collision risk.

[0081] In another embodiment, the length set includes the length of the first parking trajectory line, the length of the second parking trajectory line, and the parking drivable length planned by the planning module.

[0082] In other embodiments of this application, the minimum length between the length of the first parking trajectory line and the parking drivable length planned by the planning module can be determined as the parking drivable distance of the vehicle.

[0083] After obtaining the available parking distance, the risk of collision can be assessed based on this distance. All other things being equal, a shorter available parking distance indicates a closer distance between the vehicle and the obstacle, increasing the risk of a collision; conversely, a longer available parking distance indicates a greater distance between the vehicle and the obstacle, decreasing the risk of a collision.

[0084] In addition to the driving distance for parking, parking-related information such as the road surface slope information (e.g., whether the slope is uphill or downhill, and the angle of the slope), the vehicle's motion status (e.g., vehicle speed), and the area information of the parking area (e.g., the size and complexity of the parking area) can also be used as factors in comprehensively assessing collision risk to support accurate assessment of collision risk.

[0085] For example, assuming all other factors remain constant, the higher the vehicle speed, the shorter the time it takes for the vehicle to hit an obstacle, and the more likely a collision risk will occur. Conversely, the lower the vehicle speed, the longer the time it takes for the vehicle to hit an obstacle, and the less likely a collision risk will occur.

[0086] To illustrate a feasible implementation of assessing the existence of collision risk based on the parking drivable distance, in one embodiment of this application, the parking control method further includes: inputting the acquired parking drivable distance and the vehicle's current parking-related information into a learned neural network, obtaining the output result of the learned neural network, and using the output result to describe whether the vehicle has a collision risk.

[0087] The learned neural network is obtained through regression learning using a normalized exponential function (or softmax function). The data used for regression learning includes the test vehicle's drivable parking distance and related parking information in parking scenarios with collision risk, as well as the test vehicle's drivable parking distance and related parking information in parking scenarios without collision risk. The parking information includes at least one of the following: road surface slope information, vehicle motion status, and area information of the parking area.

[0088] In one embodiment, test data from previous parking scenarios (with and without collision risk) can be fed back. The drivable parking distance and related parking information from the test data are used as input, and a neural network is used for softmax regression learning. The learned neural network is then used to determine whether the vehicle faces a parking collision risk. After the neural network's regression learning is complete, the vehicle's current drivable parking distance and related parking information can be used as input data to the learned neural network. Based on the output of the learned neural network, the system determines whether the vehicle faces a parking collision risk.

[0089] In one embodiment, the output of the learned neural network can be information indicating whether the vehicle is at risk of a parking collision.

[0090] In another embodiment, the output can also be information representing the probability of a parking collision for the vehicle. The presence of a parking collision risk can be determined by comparing the parking collision probability with a threshold.

[0091] By performing neural network regression learning based on test data under different parking scenarios, and inputting the currently acquired drivable parking distance and parking-related information into the learned neural network, it is possible to accurately determine whether a vehicle is at risk of a parking collision.

[0092] Step 102: If it is determined that there is no risk of collision with the vehicle based on the drivable parking distance, a parking reference trajectory is obtained based on the drivable parking distance and the vehicle's current expected speed sequence. The expected speed sequence includes the vehicle's expected speed at a series of future time points, and the parking reference trajectory includes the vehicle's longitudinal reference position and reference speed at a series of future time points.

[0093] Unlike when parking safety is prioritized when there is a collision risk, if there is no collision risk during parking, the focus is on parking stability, and a predictive function control algorithm can be used to control parking acceleration.

[0094] Predictive function control (PFC) is a control algorithm belonging to model predictive control (MMC) and possesses good stability, robustness, dynamic response characteristics, and control accuracy. While maintaining the advantages of MMC, PFC makes the generated control inputs more predictable and effectively reduces the computational load, thus adapting to the rapid control requirements of fast-responding controlled objects. By using PFC to control parking acceleration in the absence of parking collision risk, it helps to achieve stable vehicle speed control in automated parking scenarios.

[0095] Predictive function control algorithms can control parking acceleration by tracking a suitable parking reference trajectory. In this embodiment, the parking reference trajectory can be obtained based on the current drivable parking distance and the vehicle's current desired speed sequence. The parking reference trajectory is a reference trajectory for achieving longitudinal (i.e., the longitudinal axis direction of the vehicle coordinate system) control during parking. It can include the longitudinal position and speed at various points in time over a future period, and the longitudinal length of the parking reference trajectory does not exceed the drivable parking distance. This not only enables stable control of the vehicle's parking speed at extremely low speeds but also provides some control over the parking position distance.

[0096] In one embodiment, the desired speed sequence can be the vehicle's desired speed at various points in time (e.g., within 2 seconds) planned by the vehicle's planning module, and the parking reference trajectory includes the reference speeds of each vehicle, which constitutes the desired speed sequence. The longitudinal reference positions of each vehicle included in the parking reference trajectory can be obtained based on the vehicle's current motion state, the desired speed sequence, and the interval between adjacent time points. By tracking the parking reference trajectory to control the vehicle's parking acceleration, stable control of the parking speed can be achieved, thereby supporting longitudinal driving stability during extremely low-speed parking.

[0097] In one embodiment of this application, obtaining a parking reference trajectory based on the parking drivable distance and the vehicle's current expected speed sequence includes: if the second parking trajectory line has a minimum length, obtaining the vehicle's longitudinal reference position based on the vehicle's longitudinal position in the second parking trajectory line.

[0098] Since the second parking trajectory line has a minimum length, it can be assumed that there are parking obstacles that affect the vehicle's parking but do not pose a collision risk at the moment. If it is expected that the parking trajectory line does not involve avoiding the parking obstacle in the longitudinal direction, then the longitudinal reference position of the vehicle can be obtained based on the longitudinal position of the vehicle in the second parking trajectory line, so as to ensure that the vehicle can continue to park for a period of time afterward without colliding with the parking obstacle.

[0099] In one embodiment of this application, obtaining a parking reference trajectory based on the parking drivable distance and the vehicle's current expected speed sequence includes: if the first parking trajectory line has a minimum length, obtaining the vehicle's longitudinal reference position based on the vehicle's longitudinal position in the first parking trajectory line.

[0100] Since the first parking trajectory line has the minimum length, it can be assumed that there are no parking obstacles that affect the parking of the vehicle. Therefore, the longitudinal reference position of the vehicle can be obtained based on the longitudinal position of the vehicle in the first parking trajectory line, so as to ensure that the vehicle can continue to park for a period of time and usually will not collide with obstacles.

[0101] In one embodiment of this application, obtaining a parking reference trajectory based on the drivable parking distance and the vehicle's current expected speed sequence includes: if the drivable parking length is the minimum length, obtaining the vehicle's longitudinal reference position based on the expected speed sequence output by the vehicle's planning module and the vehicle's current motion state.

[0102] Since the drivable length for parking is the minimum length, it can be assumed that the planning module senses parking obstacles that affect the vehicle's parking but do not pose a collision risk at the moment. If the desired parking trajectory line does not involve avoiding the parking obstacle in the longitudinal direction, the longitudinal reference position of the vehicle can be obtained based on the vehicle's current motion state (such as the vehicle's current speed and acceleration) and the planned desired speed sequence, so as to ensure that the vehicle can continue to park for a period of time afterward without colliding with the parking obstacle.

[0103] Step 103: Using a predictive function control algorithm, obtain the first acceleration for controlling the vehicle's parking based on the vehicle's current motion state and parking reference trajectory.

[0104] Predictive function control algorithms can constrain the parking acceleration control input as a linear combination of basis functions with a fixed structure. The basis functions can be step functions, which means that the control input is predicted to remain consistent at each time step.

[0105] The prediction model in the predictive function control algorithm can be a second-order vehicle model with respect to the vehicle's longitudinal position and velocity. The prediction model can then predict the vehicle's longitudinal parking trajectory based on the vehicle's current motion state and the consistency constraint on acceleration imposed by fixed basis functions.

[0106] The predictive function control algorithm can track a suitable reference trajectory to achieve longitudinal control during parking. Therefore, the reference trajectory tracked by the longitudinal control can be a time series of longitudinal position and velocity. Feasibly, the tracked reference trajectory can be the parking reference trajectory described above.

[0107] In the predictive function control algorithm, the optimization objective can be to find the predicted output that, under a fixed basis function, approximates the reference trajectory as closely as possible within the optimization time domain. If a quadratic performance index is used, the acceleration request can be solved by minimizing the sum of squared errors between the reference trajectory and the predicted output within the optimization time domain. Feasibly, the predicted output of the prediction model can be the predicted parking trajectory, and the solved acceleration request can be the aforementioned first acceleration.

[0108] When a vehicle is parked at extremely low speeds, it is easily affected by factors such as road slope and road friction, which can cause a certain deviation between the prediction model's output and the vehicle's actual execution. Therefore, the model parameters of the prediction model can be optimized based on the error between the prediction model's output and the vehicle's actual execution.

[0109] In one embodiment, the prediction model parameters of the prediction function control algorithm can be optimized based on the error between the vehicle's parking reference trajectory and the actual control result. The acceleration output by the parameter-optimized prediction model can then be used to achieve vehicle parking control. This can help eliminate the impact of disturbances during low-speed vehicle driving on the vehicle's parking stability, ensuring that the actual vehicle control conforms to the parking reference trajectory.

[0110] In one embodiment of this application, the algorithm implementation based on the predictive function control algorithm may include step 103 as follows: inputting the current motion state of the vehicle into the predictive model to obtain the first acceleration output by the predictive model under the constraints of the basis function and the optimization objective.

[0111] The optimization objectives include ensuring that the error between the predicted parking trajectory and the parking reference trajectory under the first acceleration is not greater than the error between the predicted parking trajectory and the parking reference trajectory under other accelerations; the predicted parking trajectory includes the predicted longitudinal position and predicted speed of the vehicle at a series of future time points; the basis function is used to ensure that the acceleration remains consistent at a series of future time points; and the prediction model is a second-order vehicle model with respect to the longitudinal position and speed of the vehicle.

[0112] Based on the vehicle's current motion state and parking reference trajectory, the predictive model can predict the parking acceleration that meets the optimization objective. By controlling the vehicle's parking speed according to the predicted parking acceleration, it can achieve stable and safe parking by tracking the parking reference trajectory.

[0113] Vehicles are prone to errors and fluctuations when parking at extremely low speeds. Factors such as road gradient, road friction, and environmental conditions can cause these errors and fluctuations, negatively impacting parking stability. This application addresses the characteristics of predictive function control algorithms in handling these errors and fluctuations. In the absence of parking collision risk, this embodiment uses a predictive function control algorithm, combined with the vehicle's current actual motion state and parking reference trajectory, to achieve stable speed control in automatic parking scenarios.

[0114] Step 104: Control the vehicle to park based on the first acceleration.

[0115] After obtaining the acceleration required for vehicle parking, one implementation method involves determining the motor torque for parking based on the conversion relationship between acceleration and motor torque, and then using the motor torque to control vehicle parking or braking. Alternatively, the motor torque can be sent to the vehicle chassis to enable the chassis to perform parking or braking accordingly.

[0116] In another feasible implementation, if the obtained acceleration is negative, the acceleration can be converted into brake hydraulic pressure, and hydraulic calipers can be used to control the vehicle's braking accordingly.

[0117] After step 104, the step of obtaining the vehicle's parking distance can be executed again to obtain the vehicle's parking distance at the next moment. The obtained data can be used for the next execution of the parking control process until the vehicle is parked at the parking end position in the parking area.

[0118] This application applies the predictive function control algorithm to the vehicle parking control scenario. The predictive function control algorithm can make the vehicle parking acceleration more regular. Thus, by using the acceleration output by the predictive function control algorithm to control the vehicle parking in the absence of collision risk, stable control of vehicle speed can be achieved in the automatic parking scenario, which helps to achieve the stability and safety of vehicle parking.

[0119] like Figure 4As shown in the figure, this application embodiment provides a parking control device 400, including: a first acquisition module 401, used to acquire the vehicle's drivable parking distance based on the vehicle's current motion state and perception information of parking obstacles; a second acquisition module 402, used to acquire a parking reference trajectory based on the drivable parking distance and the vehicle's current expected speed sequence when it is determined that there is no collision risk based on the drivable parking distance, wherein the expected speed sequence includes the vehicle's expected speed at a series of future time points, and the parking reference trajectory includes the vehicle's longitudinal reference position and reference speed at a series of future time points; a third acquisition module 403, used to acquire a first acceleration for controlling the vehicle's parking based on the vehicle's current motion state and parking reference trajectory using a predictive function control algorithm; and a control module 404, used to control the vehicle's parking based on the first acceleration.

[0120] One embodiment of this application provides an electronic chip, including: a processor for executing computer program instructions stored in a memory, wherein when the computer program instructions are executed by the processor, the electronic chip is triggered to execute the method described in any embodiment of this application.

[0121] One embodiment of this application provides an electronic device including at least one processor and a memory coupled together. The memory is used to store computer program instructions, and the processor is used to execute the computer program instructions. When the computer program instructions are executed by the processor, the electronic device is triggered to execute the method described in any embodiment of this application.

[0122] One embodiment of this application provides a computer-readable storage medium storing a computer program that, when run on a computer, causes the computer to perform the methods described in any embodiment of this application.

[0123] One embodiment of this application provides a computer program product, which includes a computer program that, when run on a computer, causes the computer to perform the methods described in any embodiment of this application.

[0124] Figure 5 This is a schematic diagram of a computer device provided according to one embodiment of this application. Figure 5 As shown, the computer device 20 in this embodiment includes a processor 21 and a memory 22. The memory 22 stores a computer program 23 that can run on the processor 21. When the computer program 23 is executed by the processor 21, it implements the steps in the method embodiments of this application. To avoid repetition, these steps are not described in detail here. Alternatively, when the computer program 23 is executed by the processor 21, it implements the functions of each model / unit in the device embodiments of this application. To avoid repetition, these functions are not described in detail here.

[0125] Computer device 20 includes, but is not limited to, processor 21 and memory 22. Those skilled in the art will understand that... Figure 5 This is merely an example of computer device 20 and does not constitute a limitation on computer device 20. It may include more or fewer components than shown, or combine certain components, or different components. For example, computer device may also include input / output devices, network access devices, buses, etc.

[0126] The processor 21 can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor, or it can be any conventional processor.

[0127] The memory 22 can be an internal storage unit of the computer device 20, such as a hard disk or RAM of the computer device 20. The memory 22 can also be an external storage device of the computer device 20, such as a plug-in hard disk, Smart Media (SM) card, Secure Digital (SD) card, or FlashCard equipped on the computer device 20. Furthermore, the memory 22 can include both internal and external storage units of the computer device 20. The memory 22 is used to store the computer program 23 and other programs and data required by the computer device. The memory 22 can also be used to temporarily store data that has been output or will be output.

[0128] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, apparatuses, or units, and may be electrical, mechanical, or other forms.

[0129] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0130] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in a combination of hardware and software functional units.

[0131] An integrated unit implemented as a software functional unit can be stored in a computer-readable storage medium. This software functional unit, stored in a storage medium, includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) or processor to execute some steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, portable hard drive, read-only memory (ROM), random access memory (RAM), magnetic disk, or optical disk.

[0132] In this application, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0133] This application can be described in the general context of computer-executable instructions, such as program modules, that are executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform a specific task or implement a specific abstract data type. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0134] Those skilled in the art will recognize that the units and algorithm steps described in the embodiments of this application can be implemented using electronic hardware, computer software, or a combination of electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0135] Those skilled in the art will readily understand that, for the sake of convenience and brevity, the same or similar parts between the various embodiments of this application can be referred to mutually. For example, the specific working processes of the systems, devices, and units described in the embodiments of this application can be referred to the corresponding processes in the method embodiments of this application, and will not be repeated here.

[0136] The above description is merely a specific embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A parking control method, characterized in that, include: Based on the vehicle's current motion state and the perception information of parking obstacles, the vehicle's parking driving distance is obtained; If it is determined that there is no risk of collision with the vehicle based on the available parking distance, a parking reference trajectory is obtained based on the available parking distance and the current expected speed sequence of the vehicle. The expected speed sequence includes the expected speed of the vehicle at a series of future time points, and the parking reference trajectory includes the longitudinal reference position and reference speed of the vehicle at a series of future time points. Using a predictive function control algorithm, a first acceleration for controlling the parking of the vehicle is obtained based on the current motion state of the vehicle and the parking reference trajectory; The vehicle is parked based on the first acceleration.

2. The method according to claim 1, characterized in that, The predictive function control algorithm, based on the vehicle's current motion state and the parking reference trajectory, obtains a first acceleration for controlling the vehicle's parking, including: The current motion state of the vehicle is input into the prediction model to obtain the first acceleration output by the prediction model under the constraints of the basis function and the optimization objective; The optimization objective includes that the error between the predicted parking trajectory of the vehicle under the first acceleration and the parking reference trajectory is not greater than the error between the predicted parking trajectory of the vehicle under other accelerations and the parking reference trajectory. The predicted parking trajectory includes the predicted longitudinal position and predicted speed of the vehicle at a series of future time points. The basis functions are used to ensure that the acceleration remains consistent across a series of future time points, and the prediction model is a second-order vehicle model with respect to the vehicle's longitudinal position and vehicle speed.

3. The method according to claim 1, characterized in that, The step of obtaining the vehicle's parking distance based on the vehicle's current motion state and perception information of parking obstacles includes: Based on the vehicle's current motion state, the vehicle's current turning radius is obtained; Based on the turning radius, a first parking trajectory line is obtained, wherein the radius of curvature of the first parking trajectory line is the turning radius, the starting point of the first parking trajectory line is the current position of the vehicle, and the longitudinal position of the ending point of the first parking trajectory line is the longitudinal position of the parking end position of the desired parking trajectory line planned by the vehicle's planning module. Based on the turning radius and the perception information of parking obstacles, a second parking trajectory line is obtained, wherein the radius of curvature of the second parking trajectory line is the turning radius, the starting point of the second parking trajectory line is the current position of the vehicle, and if the perception information indicates that the vehicle perceives a parking obstacle, then the ending point of the second parking trajectory line is the intersection of the perceived parking obstacle and the second parking trajectory line. Obtain a length set, the length set including the length of the first parking trajectory line and the length of the second parking trajectory line; The minimum length in the set of lengths is determined as the parking distance of the vehicle.

4. The method according to claim 3, characterized in that, The second parking trajectory line includes parking trajectory lines for multiple corner points of the vehicle.

5. The method according to claim 3, characterized in that, The step of obtaining the parking reference trajectory based on the parking drivable distance and the vehicle's current expected speed sequence includes: If the second parking trajectory line has the minimum length, the longitudinal reference position of the vehicle is obtained based on the longitudinal position of the vehicle in the second parking trajectory line.

6. The method according to claim 3, characterized in that, If the perception information indicates that the vehicle has not detected a parking obstacle, the length of the second parking trajectory line is greater than the length of the first parking trajectory line; The step of obtaining the parking reference trajectory based on the parking drivable distance and the vehicle's current expected speed sequence includes: If the first parking trajectory line has the minimum length, the longitudinal reference position of the vehicle is obtained based on the longitudinal position of the vehicle in the first parking trajectory line.

7. The method according to claim 3, characterized in that, The length set also includes: the drivable length for parking, which is planned by the vehicle's planning module based on the perception information of parking obstacles; The step of obtaining the parking reference trajectory based on the parking drivable distance and the vehicle's current expected speed sequence includes: If the parking drivable length is the minimum length, the longitudinal reference position of the vehicle is obtained based on the expected speed sequence output by the vehicle's planning module and the vehicle's current motion state.

8. The method according to any one of claims 1-7, characterized in that, The method further includes: The acquired drivable parking distance and the vehicle's current parking information are input into the learned neural network to obtain the output of the learned neural network. The output is used to describe whether the vehicle has a collision risk. The learned neural network is obtained by using a normalized exponential function regression learning method. The data used for regression learning includes the driving distance and parking-related information of the test vehicle in a parking scenario with collision risk, and the driving distance and parking-related information of the test vehicle in a parking scenario without collision risk. The parking-related information includes at least one of the following: road surface slope information of the road where the vehicle is located, vehicle movement status, and area information of the parking area.

9. The method according to any one of claims 1-7, characterized in that, The method further includes: applying brakes to the vehicle if a collision risk is determined based on the parking distance.

10. A parking control device, characterized in that, include: The first acquisition module is used to acquire the parking distance of the vehicle based on the vehicle's current motion state and the perception information of parking obstacles. The second acquisition module is used to acquire a parking reference trajectory based on the parking drivable distance and the current expected speed sequence of the vehicle when it is determined that there is no collision risk to the vehicle based on the parking drivable distance. The expected speed sequence includes the expected speed of the vehicle at a series of future time points, and the parking reference trajectory includes the longitudinal reference position and reference speed of the vehicle at a series of future time points. The third acquisition module is used to acquire a first acceleration for controlling the parking of the vehicle based on the current motion state of the vehicle and the parking reference trajectory using a prediction function control algorithm. A control module is used to control the vehicle to park based on the first acceleration.

11. An electronic device, characterized in that, The electronic device includes at least one processor coupled to a memory for storing computer program instructions and for executing the computer program instructions, wherein when the computer program instructions are executed by the processor, the electronic device is triggered to perform the method as described in any one of claims 1-9.

12. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when run on a computer, causes the computer to perform the method as described in any one of claims 1-9.

Images