Acceleration control method and device in parking stage, storage medium and electronic equipment

Abstract

The invention provides an acceleration control method and device in a parking stage, a storage medium and electronic equipment. The method comprises the following steps: acquiring a front acceleration, a vehicle speed and an obstacle distance corresponding to a current sampling time point; determining a planned speed and a planned distance based on the vehicle speed and the obstacle distance; determining an expected end acceleration; determining an expected average acceleration according to the planned speed, the planned distance, the front acceleration and the expected tail end acceleration; determining a planned jerk; determining Bezier curve data of the current sampling time point according to the expected average acceleration, the preposed acceleration, the planned jerk acceleration and the expected end acceleration, and determining a curve planned acceleration based on the Bezier curve data; and carrying out slope limiting processing on the curve planning acceleration, and taking a processing result as the current request acceleration. By applying the method provided by the invention, acceleration planning is realized based on the Bezier curve, so that acceleration change is smooth, and user experience is improved.

Description

Technical Field

[0001] The present invention relates to the field of vehicle control technology, and in particular to an acceleration control method and device, a storage medium and an electronic device during a parking phase. Background Art

[0002] With the development of autonomous driving technology, autonomous vehicles, i.e. vehicles equipped with autonomous driving systems, have gradually entered people's daily lives. In the process of autonomous driving, parking scenarios are one of the main driving scenarios. For example, when the vehicle in front of the vehicle stops or other obstacles appear in front of the vehicle, the vehicle needs to be controlled to stop at a safe distance.

[0003] At present, the acceleration control method of the parking stage of an autonomous vehicle is mainly implemented by uniform deceleration control plus constant deceleration control. Specifically, when the vehicle enters the control state of the parking stage, the acceleration required for the vehicle to stop before a certain safe distance is calculated by the uniform deceleration formula based on the current speed of the vehicle, the relative distance between the vehicle and the vehicle in front or obstacles, and other parameters. The acceleration is used to decelerate the vehicle. When the speed of the vehicle is less than the preset threshold, a preset constant value is used as the acceleration, and the vehicle is decelerated using the preset value until it stops. The acceleration applied to the vehicle for deceleration control is also called deceleration.

[0004] The inventors have discovered through research that in the existing acceleration control method, there is usually a certain difference in deceleration between the uniform deceleration control stage and the constant deceleration control stage. When the deceleration changes, it is easy to cause the vehicle to experience poor physical sensations such as a sense of frustration, sudden braking, and vehicle shaking, resulting in a poor user experience. Summary of the invention

[0005] In view of this, an embodiment of the present invention provides an acceleration control method during the parking phase to solve the problem that in the existing acceleration control process, the vehicle acceleration is prone to instantaneous and obvious changes, resulting in physical problems such as vehicle jerking, resulting in a poor driving experience for users.

[0006] The embodiment of the present invention further provides an acceleration control device during the parking phase to ensure the practical implementation and application of the above method.

[0007] To achieve the above objectives, the embodiments of the present invention provide the following technical solutions:

[0008] A first aspect of an embodiment of the present invention provides an acceleration control method during a parking phase, comprising:

[0009] When the controlled vehicle is in the parking stage, the pre-acceleration, vehicle speed and obstacle distance corresponding to the current sampling time point are obtained; the pre-acceleration is the requested acceleration corresponding to the previous sampling time point of the current sampling time point, and the obstacle distance is the current distance between the controlled vehicle and the obstacle;

[0010] Determine the planned speed and planned distance corresponding to the current sampling time point based on the vehicle speed and the obstacle distance;

[0011] Determining a desired terminal acceleration corresponding to the controlled vehicle;

[0012] Determining an expected average acceleration corresponding to the current sampling time point according to the planned speed, the planned distance, the leading acceleration and the expected terminal acceleration;

[0013] Determine the planned acceleration corresponding to the current sampling time point;

[0014] Determine the Bezier curve data corresponding to the current sampling time point according to the expected average acceleration, the leading acceleration, the planned jerk and the expected terminal acceleration, and determine the curve planned acceleration corresponding to the current sampling time point based on the Bezier curve data; the Bezier curve data represents a Bezier curve describing the acceleration of the vehicle during the parking phase;

[0015] The curve planning acceleration is subjected to slope limiting processing to obtain a target acceleration corresponding to the curve planning acceleration, and the target acceleration is used as the requested acceleration corresponding to the current sampling time point to control the acceleration of the controlled vehicle.

[0016] A second aspect of an embodiment of the present invention provides an acceleration control device in a parking phase, comprising:

[0017] an acquisition unit, for acquiring, when the controlled vehicle is in a parking phase, a pre-acceleration, a vehicle speed, and an obstacle distance corresponding to a current sampling time point; the pre-acceleration is a requested acceleration corresponding to a sampling time point before the current sampling time point, and the obstacle distance is a distance between the controlled vehicle and the obstacle at present;

[0018] A first determining unit, configured to determine a planned speed and a planned distance corresponding to the current sampling time point based on the vehicle speed and the obstacle distance;

[0019] A second determination unit, used to determine the desired terminal acceleration corresponding to the controlled vehicle;

[0020] A third determining unit, configured to determine an expected average acceleration corresponding to the current sampling time point according to the planned speed, the planned distance, the leading acceleration and the expected terminal acceleration;

[0021] A fourth determining unit, configured to determine a planned jerk corresponding to the current sampling time point;

[0022] a fifth determining unit, configured to determine Bezier curve data corresponding to the current sampling time point according to the expected average acceleration, the leading acceleration, the planned jerk, and the expected terminal acceleration, and determine the curve planned acceleration corresponding to the current sampling time point based on the Bezier curve data; the Bezier curve data represents a Bezier curve describing the acceleration of the vehicle during the parking phase;

[0023] The slope limiting unit is used to perform slope limiting processing on the curve planning acceleration to obtain a target acceleration corresponding to the curve planning acceleration, and use the target acceleration as the requested acceleration corresponding to the current sampling time point to control the acceleration of the controlled vehicle.

[0024] A third aspect of an embodiment of the present invention provides a storage medium, which includes stored instructions, wherein when the instructions are executed, the device where the storage medium is located is controlled to execute the acceleration control method in the parking phase as described above.

[0025] A fourth aspect of an embodiment of the present invention provides an electronic device, comprising a memory, and one or more instructions, wherein the one or more instructions are stored in the memory and are configured to be executed by one or more processors to perform the acceleration control method in the parking phase as described above.

[0026] Based on the above-mentioned embodiment of the present invention, an acceleration control method for a parking phase is provided, comprising: when a controlled vehicle is in the parking phase, obtaining a pre-acceleration, a vehicle speed and an obstacle distance corresponding to a current sampling time point; determining a planned speed and a planned distance corresponding to the current sampling time point based on the vehicle speed and the obstacle distance; determining an expected terminal acceleration corresponding to the controlled vehicle; determining an expected average acceleration corresponding to the current sampling time point according to the planned speed, the planned distance, the pre-acceleration and the expected terminal acceleration; determining a planned jerk corresponding to the current sampling time point; determining Bezier curve data corresponding to the current sampling time point according to the expected average acceleration, the pre-acceleration, the planned jerk and the expected terminal acceleration, and determining a planned acceleration of the curve corresponding to the current sampling time point based on the Bezier curve data; performing slope limiting processing on the planned acceleration of the curve to obtain a target acceleration corresponding to the planned acceleration of the curve, and using the target acceleration as the requested acceleration corresponding to the current sampling time point to control the acceleration of the controlled vehicle. By applying the method provided in the embodiment of the present invention, the vehicle state parameters such as the current front acceleration, expected average acceleration, planned acceleration and expected terminal acceleration can be determined at each sampling time point, the corresponding Bezier curve data under the current state of the vehicle can be solved, acceleration planning can be performed based on the Bezier curve data, and then the current requested acceleration can be determined through slope limitation. The Bezier curve has a smooth characteristic, and the corresponding acceleration described by the Bezier curve changes smoothly. By using the Bezier curve for acceleration planning, the front and rear accelerations can be smoothly connected, so that the acceleration of the vehicle can smoothly transition to the expected terminal acceleration, and the final requested acceleration can be obtained through slope limitation processing, which can further ensure the smoothness of the acceleration change, reduce the adverse physical sensation caused by the acceleration change, and improve the user's driving experience and riding experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0028] Figure 1 A flow chart of an acceleration control method during a parking phase provided by an embodiment of the present invention;

[0029] Figure 2 A schematic diagram of a velocity square-displacement coordinate system provided by an embodiment of the present invention;

[0030] Figure 3 A schematic diagram of a Bezier curve describing acceleration provided by an embodiment of the present invention;

[0031] Figure 4 A flow chart of an acceleration control method during a parking phase provided by an embodiment of the present invention;

[0032] Figure 5 An overall scheme planning flow chart of an acceleration control method in a parking phase provided by an embodiment of the present invention;

[0033] Figure 6 A schematic diagram of the structure of an acceleration control device in a parking phase provided by an embodiment of the present invention;

[0034] Figure 7 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0035] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0036] In this application, the terms "comprises", "comprising" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprising a ..." does not exclude the presence of other identical elements in the process, method, article or device comprising the element.

[0037] As can be seen from the background technology, the acceleration control process of the parking stage of the automatic driving car usually first calculates an acceleration based on the uniform deceleration principle, and performs deceleration control by the acceleration until the vehicle speed is lower than the preset threshold, and then performs deceleration control with the preset acceleration. Before the acceleration is changed, the acceleration of the vehicle is calculated based on parameters such as the real-time speed and distance of the vehicle. The acceleration calculated under different initial vehicle states is different. After the acceleration is changed, the acceleration of the vehicle is a preset constant value. Therefore, the acceleration before and after the acceleration change usually has a large difference, and the slope of the acceleration before and after may switch between positive and negative, which may easily cause the vehicle to feel uncomfortable during driving, such as jerkiness, violent braking, and vehicle shaking, which will cause adverse effects on the user's driving experience and riding experience. Therefore, the embodiment of the present invention provides an acceleration control method for the parking stage, which performs acceleration planning based on the Bezier curve at the sampling time point, adjusts the vehicle acceleration in real time, and makes the vehicle acceleration smoothly transition to the desired terminal acceleration, avoiding the jerkiness, violent braking, and vehicle shaking caused by the acceleration change, so as to improve the user experience.

[0038] An embodiment of the present invention provides an acceleration control method during parking. The method can be applied to an autonomous vehicle. The execution subject can be a vehicle controller. The method flow chart of the method is as follows: Figure 1 As shown, including:

[0039] S101: When the controlled vehicle is in a parking stage, obtaining the pre-acceleration, vehicle speed and obstacle distance corresponding to the current sampling time point; the pre-acceleration is the requested acceleration corresponding to the previous sampling time point of the current sampling time point, and the obstacle distance is the current distance between the controlled vehicle and the obstacle;

[0040] In the method provided by the embodiment of the present invention, the controlled vehicle can be an automatic driving car. During the operation of the controlled vehicle, the controller will sample the state parameters of the vehicle in real time according to a preset sampling period, and determine the current requested acceleration according to the current state parameters and the current acceleration planning strategy, and send the requested acceleration to the actuator of the vehicle, so that the actuator adjusts the acceleration of the vehicle to the requested acceleration, thereby controlling the acceleration of the vehicle and then controlling the speed of the vehicle. For example, the sampling period is 0.02s, that is, each interval of 0.02 seconds is a sampling time point. When the sampling time point is reached, the controller performs parameter sampling and then realizes acceleration regulation. During the automatic driving process of the controlled vehicle, if the controller detects the presence of a preset parking scene, the controlled vehicle will enter the parking stage, that is, the stage of controlling the vehicle to decelerate and stop. At this time, acceleration planning is performed according to the acceleration planning strategy of the parking stage. The parking scene may be an obstacle that blocks the passage in front. The obstacle may be a non-vehicle object such as a vehicle or pedestrian. For example, when the vehicle in front is stationary and close to the controlled vehicle, the vehicle in front can be regarded as an obstacle. For example, when both the vehicle in front and the controlled vehicle are driving at a low speed and the distance between the two vehicles is close, the vehicle in front can also be regarded as an obstacle. After the controlled vehicle enters the parking stage, when each sampling time point is reached, the controller can obtain the current speed of the vehicle and the current distance between the vehicle and the obstacle in front through the vehicle sensor, and use the current speed as the vehicle speed corresponding to the current sampling time point, and use the distance between the vehicle and the obstacle as the obstacle distance corresponding to the current sampling time point. From the stored acceleration data, obtain the requested acceleration of the previous sampling time point of the current sampling time point, that is, the acceleration sent by the controller to the actuator at the previous sampling time point for acceleration control, and use this acceleration as the pre-acceleration corresponding to the current sampling time point.

[0041] S102: Determine the planned speed and planned distance corresponding to the current sampling time point based on the vehicle speed and the obstacle distance;

[0042] In the method provided by the embodiment of the present invention, the controller can determine the initial speed and initial distance currently used for acceleration planning based on the vehicle speed and obstacle distance corresponding to the current sampling time point according to the predetermined planning strategy, and use the initial speed as the planning speed at the current sampling time point, and use the initial distance as the planning distance at the current sampling time point. The planning strategy can be set as needed, for example, using the vehicle speed as the planning speed and the obstacle distance as the planning distance, or setting a compensation method in combination with the response delay of the actuator to compensate for the vehicle speed and obstacle distance to obtain the planning speed and planning distance.

[0043] S103: Determine the expected terminal acceleration corresponding to the controlled vehicle;

[0044] In the method provided in an embodiment of the present invention, the speed state of the vehicle expected at the end of the parking stage can be pre-configured in the controller. The controller can calculate the acceleration expected at the end of the parking stage, that is, the acceleration before the vehicle stops, based on the vehicle state after the controlled vehicle enters the parking stage and the speed state expected at the end of the parking stage.

[0045] S104: Determine an expected average acceleration corresponding to the current sampling time point according to the planned speed, the planned distance, the leading acceleration and the expected terminal acceleration;

[0046] In the method provided by an embodiment of the present invention, the controller can calculate the expected average acceleration corresponding to the current sampling time point based on the planned speed, planned distance, leading acceleration and expected terminal acceleration corresponding to the current sampling time point, that is, the average acceleration currently used for acceleration planning, which can be understood as the average acceleration expected to be achieved during the parking stage under the current vehicle state.

[0047] S105: Determine the planned acceleration corresponding to the current sampling time point;

[0048] In the method provided by the embodiment of the present invention, acceleration planning is performed based on the Bezier curve during the parking stage. The acceleration change curve of the parking stage is described by the Bezier curve, and the actual Bezier curve under the current vehicle state is calculated at each sampling time point to obtain an acceleration on the Bezier curve that matches the current time point. After further slope restriction is imposed on the acceleration, the current requested acceleration is obtained. In the acceleration planning process of the parking stage, it is necessary to determine the jerk corresponding to the current sampling time point in the coordinate system of the Bezier curve, and use the jerk as the planned jerk at the current sampling time point for acceleration planning. The jerk is the slope of the acceleration, which is represented by the parameter Jerk.

[0049] S106: Determine Bezier curve data corresponding to the current sampling time point according to the expected average acceleration, the leading acceleration, the planned jerk, and the expected terminal acceleration, and determine the curve planned acceleration corresponding to the current sampling time point based on the Bezier curve data; the Bezier curve data represents a Bezier curve describing the acceleration of the vehicle during the parking phase;

[0050] In the method provided by the embodiment of the present invention, the expected average acceleration, lead acceleration, planned jerk and expected terminal acceleration corresponding to the current sampling time point are used to characterize the vehicle state corresponding to the current sampling time point, so as to calculate the actual Bezier curve under the current vehicle state. By solving the interpolation points on the actual Bezier curve, the acceleration value corresponding to the current sampling time point on the Bezier curve can be obtained, and the acceleration value is used as the planned acceleration of the curve corresponding to the current sampling time point. Specifically, the basic parameters of the Bezier curve describing the acceleration of the parking stage, such as the order, the number of segments, etc., can be pre-set, and then the Bezier curve can be solved based on the current vehicle state parameters.

[0051] In order to better illustrate the method provided by the embodiment of the present invention, a brief description of the vehicle acceleration using a Bezier curve is given here. 2 The (speed squared)-S (displacement) coordinate system describes the speed curve of the vehicle during the parking phase, e.g. Figure 2 As shown, V0 represents the initial vehicle speed of the plan, V n Represents the vehicle speed at the planned end point. As shown in the following formulas (1) to (3), based on the axiom formula (1), formula (2) can be calculated, and then formula (3) can be derived. It is proved that the definite integral of the acceleration of the Bezier curve in the t coordinate system is the average deceleration of the entire acceleration curve in the time coordinate system. In formulas (1) to (3), v n Equivalent to V n , v0 is equivalent to V0, v i represents the vehicle speed corresponding to position i, and Δs is a minimum value of a distance. i represents the longitudinal acceleration of the vehicle at position i. S is the distance to be moved in the entire planning, and s is the longitudinal distance variable of the movement.

[0052]

[0053] Set the Bezier curve formula. The order of the Bezier curve is represented by parameter n. For example, n can be set to 4. The number of segments of the Bezier curve is represented by seg. For example, seg can be set to 6. As shown in formula (4), the control point coefficient (i.e., basis function) ax of the Bezier curve can be calculated based on formula (4): i+1 (t), where C represents the permutation and combination formula, and t is the auxiliary parameter t in the Bezier curve. In the scenario of acceleration planning in the parking phase, t can be regarded as the percentage of the planned remaining distance. In the acceleration control process, based on the real-time vehicle state parameters, the Bezier curve can be used to describe the vehicle from the current position to the target terminal parking position, and the remaining distance percentage t is used as the acceleration curve of the coordinate system. According to v 2 Divide the Bezier curve describing the acceleration into seg segments. The example diagram of the Bezier curve can be shown as follows Figure 3As shown, the vertical coordinate is the acceleration ax. Each curve has n+1 control points. Figure 3 The control points in each curve segment are marked with graphics of different shapes, such as Figure 3 The circles marked on the curve, the cross symbols formed by different numbers of line segments, the squares and the diamonds. The value of t in the Bezier curve corresponds to the position of the vehicle. For example, when t=0, the vehicle is at the current position. When t=1, the vehicle is at the end position of the first Bezier curve. When t=seg, the vehicle is at the preset tgtdist position, and the vehicle speed is the preset tgtspeed. tgtdist represents the end distance of the parking stage, that is, the distance between the vehicle and the obstacle in front before the vehicle stops (the end of the parking stage). For example, tgtdist can be set to 3.08m. tgtspeed represents the end speed of the parking stage, that is, the vehicle speed before the vehicle stops (the end of the parking stage). For example, it can be set to 1kph. As shown in the following formulas, based on formula (4), the curve parameter Ax representing the acceleration of a single-segment Bezier curve can be obtained. By calculating the corresponding first and second differential terms and integral terms through formulas (5) to (7), the Jerk curve parameter, Snap curve parameter and speed square increment curve parameter v2 representing the single-segment Bezier curve can be obtained respectively. base Jerk is the acceleration, which indicates the slope of acceleration. The larger the value of this parameter is, the more uncomfortable the user will feel. Snap is the acceleration, which indicates the slope of acceleration. The larger the value of this parameter is, the more likely the actuator will overshoot, which will also have an adverse effect on the user's feeling.

[0054]

[0055] Ax(t)=[ax1,ax2,…ax n+1 ]

[0056]

[0057] v2 base (t) = ∫Ax dt (7)

[0058] S107: performing slope limiting processing on the curve planning acceleration to obtain a target acceleration corresponding to the curve planning acceleration, and using the target acceleration as the requested acceleration corresponding to the current sampling time point to control the acceleration of the controlled vehicle.

[0059] In the method provided by the embodiment of the present invention, the controller performs slope limiting processing on the curve planning acceleration determined by the Bezier curve data so that the slope of the requested acceleration does not exceed the limit, thereby ensuring that the change amplitude of the requested acceleration at two sampling time points is not too large. The acceleration obtained after the slope limiting processing is used as the requested acceleration at the current sampling time point, and the requested acceleration is sent to the actuator so that the actuator adjusts the vehicle acceleration to the requested acceleration to achieve acceleration control.

[0060] By applying the method provided in the embodiment of the present invention, the vehicle state parameters such as the current front acceleration, expected average acceleration, planned acceleration and expected terminal acceleration can be determined at each sampling time point, the corresponding Bezier curve data under the current state of the vehicle can be solved, acceleration planning can be performed based on the Bezier curve data, and then the current requested acceleration can be determined through slope limitation. The Bezier curve has a smooth characteristic, and the corresponding acceleration described by the Bezier curve changes smoothly. By using the Bezier curve for acceleration planning, the front and rear accelerations can be smoothly connected, so that the acceleration of the vehicle can smoothly transition to the expected terminal acceleration, and the final requested acceleration can be obtained through slope limitation processing, which can further ensure the smoothness of the acceleration change, reduce the adverse physical sensation caused by the acceleration change, and improve the user's driving experience and riding experience.

[0061] exist Figure 1 On the basis of the method shown, in the method provided by the embodiment of the present invention, the process of determining the planned speed and planned distance corresponding to the current sampling time point based on the vehicle speed and the obstacle distance mentioned in step S102 includes:

[0062] Performing speed time-delay compensation processing on the vehicle speed to obtain a time-delay estimated speed corresponding to the vehicle speed, and using the time-delay estimated speed as the planned speed;

[0063] In the method provided by the embodiment of the present invention, the speed lag compensation strategy and the distance lag compensation strategy can be set in advance according to the response delay effect of the actuator. During the acceleration control process in the parking stage, the controller performs speed lag compensation on the vehicle speed corresponding to the current sampling time point according to the predetermined speed lag compensation strategy. The speed after lag compensation is the lag estimated speed, which is used as the planned speed corresponding to the current sampling time point. Specifically, the estimated lag time of the actuator can be set in advance to obtain the requested acceleration at the lag time point. The lag time point refers to the time point before the current sampling time point, and the time interval with the current sampling time point is the lag time. The lag compensation amount of the speed is calculated by combining the requested acceleration at the lag time point, the lead acceleration and the lag time. The lag compensation amount is added to the vehicle speed corresponding to the current sampling time point to obtain the lag estimated speed.

[0064] A distance time delay compensation process is performed on the obstacle distance to obtain a time delay estimated distance corresponding to the obstacle distance, and the time delay estimated distance is used as the planned distance.

[0065] In the method provided by the embodiment of the present invention, the controller performs distance delay compensation on the obstacle distance corresponding to the current sampling time point according to a predetermined distance delay compensation strategy, and the distance after delay compensation is the delay estimated distance, and the delay estimated distance is used as the planned distance corresponding to the current sampling time point. Specifically, the delay compensation amount of the distance can be calculated based on parameters such as vehicle speed, delay time, requested acceleration at the delay time point, and pre-acceleration, and the delay compensation amount is subtracted from the obstacle distance corresponding to the current sampling time point to obtain the delay estimated distance.

[0066] Based on the method provided by the embodiment of the present invention, the vehicle speed and obstacle distance at the current sampling time point are respectively compensated for the time lag, so as to determine the planned speed and planned distance for acceleration planning. The speed and distance used in the acceleration planning take into account the response delay of the actuator, which can avoid the response overshoot caused by the large change in the slope of the acceleration change, and is conducive to reducing the control error caused by the actuator response lag, improving the stability and accuracy of the acceleration control, and can further ensure the smooth adjustment of the acceleration and avoid the problem of poor physical sensation.

[0067] exist Figure 1 On the basis of the method shown, in the method provided by the embodiment of the present invention, the process of determining the expected terminal acceleration corresponding to the controlled vehicle mentioned in step S103 includes:

[0068] If the current sampling time point is the first sampling time point when the controlled vehicle enters the parking phase, the estimated terminal acceleration is determined according to the preset parking phase terminal speed, the planned speed, the planned distance and the pre-acceleration;

[0069] In the method provided by the embodiment of the present invention, the expected terminal acceleration of the controlled vehicle is calculated according to the initial state of the vehicle entering the parking stage. It is identified whether the current sampling time point is the first sampling time point of the controlled vehicle entering the parking stage. If the current sampling time point is the first sampling time point of the parking stage, the controller calculates the terminal acceleration under the current vehicle state in a predetermined manner based on the preset parking stage terminal speed, the planned speed corresponding to the current sampling time point, the planned distance and the pre-acceleration, and uses the calculation result as the estimated terminal acceleration. The preset parking stage terminal speed is the speed that the vehicle needs to reach when it is about to park. Specifically, the parking stage terminal speed can be subtracted from the planned speed to obtain the difference between the two, the square of the difference between the two is divided by the planned distance, the calculation result is subtracted from the pre-acceleration, and the difference between the two is used as the estimated terminal acceleration.

[0070] If the estimated terminal acceleration is within a preset terminal acceleration range, the estimated terminal acceleration is used as the optimal terminal acceleration corresponding to the current sampling time point;

[0071] In the method provided in the embodiment of the present invention, based on the acceleration range of body feeling comfort, a terminal acceleration range is preset, specifically including a terminal acceleration upper limit and a terminal acceleration lower limit. The controller can compare the estimated terminal acceleration with the upper and lower limit thresholds of the terminal acceleration range respectively to determine whether the estimated terminal acceleration is within the preset terminal acceleration range. If the estimated terminal acceleration is within the terminal acceleration range, the estimated terminal acceleration is used as the optimal terminal acceleration.

[0072] If the estimated terminal acceleration is not within the terminal acceleration range, determining the optimal terminal acceleration corresponding to the current sampling time point within the range threshold of the terminal acceleration range;

[0073] In the method provided by an embodiment of the present invention, if the estimated terminal acceleration is not within the terminal acceleration range, the controller selects the optimal terminal acceleration from the range threshold of the terminal acceleration range, that is, the upper limit value and the lower limit value of the terminal acceleration. For example, if the estimated terminal acceleration is lower than the lower limit value of the terminal acceleration range, the preset lower limit value of the terminal acceleration is used as the optimal terminal acceleration. If the estimated terminal acceleration exceeds the upper limit value of the terminal acceleration range, the preset upper limit value of the terminal acceleration is used as the optimal terminal acceleration.

[0074] The optimal terminal acceleration corresponding to the current sampling time point is stored, and the optimal terminal acceleration corresponding to the current sampling time point is used as the expected terminal acceleration;

[0075] In the method provided by the embodiment of the present invention, the controller may store the optimal terminal acceleration and use it as the expected terminal acceleration.

[0076] If the current sampling time point is not the first sampling time point, the stored optimal terminal acceleration corresponding to the first sampling time point is obtained, and the optimal terminal acceleration corresponding to the first sampling time point is used as the expected terminal acceleration.

[0077] In the method provided by an embodiment of the present invention, if the current sampling time point is not the first sampling time point of the parking stage, it means that the optimal terminal acceleration has been calculated at the first sampling time point, and the controller obtains the stored optimal terminal acceleration and uses it as the expected terminal acceleration.

[0078] Based on the method provided in the embodiment of the present invention, the expected terminal acceleration is calculated based on the initial state of the vehicle entering the parking stage, and is set in combination with a pre-set terminal acceleration range. This allows the calculated expected terminal acceleration to meet the range requirements of the terminal acceleration and, on this basis, to seek the minimum slope of the overall acceleration curve, which can further ensure smooth acceleration changes.

[0079] exist Figure 1 On the basis of the method shown, in the method provided by the embodiment of the present invention, the process of determining the expected average acceleration corresponding to the current sampling time point according to the planned speed, the planned distance, the leading acceleration and the expected terminal acceleration mentioned in step S104 includes:

[0080] Determining an average acceleration range corresponding to the current sampling time point according to the planned speed, the planned distance, a preset parking phase end speed, a preset parking phase end distance, and a preset parking distance error range;

[0081] In the method provided in the embodiment of the present invention, the parking distance error range can be set in advance according to the needs, specifically including the parking distance error upper limit and the parking distance error lower limit. The controller can calculate the average acceleration upper limit based on the planned speed, planned distance, preset parking stage end distance, parking stage end speed and parking distance error upper limit corresponding to the current sampling time point, and calculate the average acceleration lower limit based on the planned speed, planned distance, parking stage end distance, parking stage end speed and parking distance error lower limit, and form an average acceleration range by the average acceleration upper limit and the average acceleration lower limit. The parking stage end speed and parking stage end distance mentioned in the embodiment of the present invention refer to the same objects as the parking stage end speed and parking stage end distance mentioned in the previous embodiment.

[0082] Determine the current planned average acceleration corresponding to the current sampling time point;

[0083] In the method provided by the embodiment of the present invention, the controller can determine the average acceleration currently used for acceleration planning, that is, the current planned average acceleration, according to the time situation of the current sampling time point. Specifically, if the current sampling time point is the first sampling time point of the parking stage, the acceleration planning of the parking stage has not been performed before. At this time, the average acceleration can be estimated based on parameters such as the end speed of the parking stage, the end distance of the parking stage, the planned speed and the planned distance, and the calculation result is used as the current planned average acceleration. If the current sampling time point is not the first sampling time point of the parking stage, the expected average acceleration corresponding to the previous sampling time point of the current sampling time point is used as the current planned average acceleration.

[0084] Determining an estimated average acceleration corresponding to the current sampling time point according to the expected terminal acceleration, the current planned average acceleration, the planned speed, the preceding acceleration and the terminal speed of the parking phase;

[0085] In the method provided in an embodiment of the present invention, the controller calculates the expected average acceleration based on parameters such as the expected terminal acceleration, the current planned average acceleration, the planned speed, the preceding acceleration and the terminal speed of the parking phase, and uses the calculation result as the estimated average acceleration at the current sampling time point.

[0086] If the estimated average acceleration is within the average acceleration range, the estimated average acceleration is used as the expected average acceleration.

[0087] In the method provided by an embodiment of the present invention, the controller may compare the estimated average acceleration with the upper and lower limits of the average acceleration range to determine whether the estimated average acceleration is within the average acceleration range. If the estimated average acceleration is within the average acceleration range, the estimated average acceleration is used as the expected average acceleration; otherwise, the expected average acceleration is selected from the upper and lower limits of the average acceleration range. For example, if the estimated average acceleration is greater than the upper limit of the average acceleration range, the upper limit of the average acceleration range is used as the expected average acceleration; if the estimated average acceleration is less than the lower limit of the average acceleration range, the lower limit of the average acceleration range is used as the expected average acceleration.

[0088] Based on the method provided in the embodiment of the present invention, the expected average acceleration used for acceleration planning can be limited, which can further ensure the smoothness of acceleration changes.

[0089] exist Figure 1 On the basis of the method shown, in the method provided by the embodiment of the present invention, the process of determining the planned acceleration corresponding to the current sampling time point mentioned in step S105 includes:

[0090] If the current sampling time point is the first sampling time point when the controlled vehicle enters the parking phase, determining the jerk corresponding to the pre-acceleration;

[0091] In the method provided by the embodiment of the present invention, when the controller determines the jerk for acceleration planning at the current sampling time point, it needs to process it according to the actual situation of the current sampling time point. If the current sampling time point is the first sampling time point when the controlled vehicle enters the parking stage, the Bezier curve has not been used for acceleration planning before. At this time, the jerk of the controlled vehicle in the time coordinate system is converted to obtain its current jerk in the Bezier curve t coordinate system. At this time, the controller needs to differentiate the pre-acceleration to obtain the jerk corresponding to the pre-acceleration.

[0092] Determine an estimated jerk corresponding to the current sampling time point based on preset Bezier curve parameters, the planned speed, the planned distance, the pre-acceleration, and the jerk corresponding to the pre-acceleration, and use the estimated jerk as the planned jerk;

[0093] In the method provided by an embodiment of the present invention, the controller calculates the acceleration in the t coordinate system based on preset Bezier curve parameters, planned distance, planned speed, precursor acceleration and the jerk corresponding to the precursor acceleration, and uses the calculation result as the estimated acceleration corresponding to the current sampling time point, and uses it as the planned acceleration.

[0094] If the current sampling time point is not the first sampling time point, the curve planning jerk corresponding to the first sampling time point is determined, and the curve planning jerk is used as the planning jerk; the first sampling time point is a sampling time point before the current sampling time point, and the curve planning jerk is an acceleration determined based on the Bezier curve data corresponding to the first sampling time point, the request acceleration corresponding to the first sampling time point, and the pre-acceleration corresponding to the first sampling time point.

[0095] In the method provided by the embodiment of the present invention, if the current sampling time point is not the first sampling time point at which the controlled vehicle enters the parking stage, the previous sampling time point of the current sampling time point is used as the first sampling time point, and the jerk determined based on the acceleration planning data corresponding to the first sampling time point is determined. Specifically, the acceleration and jerk corresponding to the first sampling time point in the t coordinate system can be calculated based on the Bezier curve data corresponding to the first sampling time point, and the jerk proportion corresponding to the current sampling time point is determined based on the acceleration and the request acceleration and pre-acceleration corresponding to the first sampling time point (i.e., the request acceleration of the previous sampling time point of the first sampling time point), and the curve planning jerk corresponding to the first sampling time point is calculated based on the jerk corresponding to the first sampling time point and the proportion corresponding to the current sampling time point. The curve planning jerk corresponding to the first sampling time point is applied as the planning jerk at the current sampling time point. The curve planning jerk corresponding to the first sampling time point can be calculated and stored in the acceleration control process of the first sampling time point, and can be directly read at the current sampling time point.

[0096] Based on the method provided in the embodiment of the present invention, the acceleration corresponding to the current sampling time point in the t coordinate system can be obtained through data conversion at the first sampling time point in the parking stage, and the acceleration on the t coordinate system can be directly calculated based on the actual Bezier curve data at subsequent sampling time points, which can ensure that acceleration planning based on the Bezier curve can be achieved at sampling time points with different data bases, thereby improving the accuracy of acceleration control.

[0097] exist Figure 1 Based on the method shown in reference Figure 4 As shown in the flowchart, in the method provided by the embodiment of the present invention, the process mentioned in step S106 of determining the Bezier curve data corresponding to the current sampling time point according to the expected average acceleration, the leading acceleration, the planned jerk and the expected terminal acceleration, and determining the curve planned acceleration corresponding to the current sampling time point based on the Bezier curve data, includes:

[0098] S201: constructing a vehicle state vector based on the expected average acceleration, the leading acceleration, the planned jerk and the expected terminal acceleration;

[0099] In the method provided by the embodiment of the present invention, the controller constructs a vector with the current expected average acceleration, the leading acceleration, the planned jerk and the expected terminal acceleration, and uses it as the vehicle state vector.

[0100] S202: solving the Bezier curve control points according to the vehicle state vector and a preset constant matrix, obtaining the control point matrix corresponding to the current sampling time point, and using the control point matrix as the Bezier curve data; the preset constant matrix is ​​a matrix constructed based on constants in the solution result matrix, and the solution result matrix is ​​a result obtained by solving the preset Bezier curve equation by Lagrange multiplier method;

[0101] In the method provided by the embodiment of the present invention, the preset Bezier curve equation is solved in advance by the Lagrange multiplier method to obtain a solution result matrix carrying unknown state parameters, the constants in the matrix are extracted, a constant matrix is ​​constructed based on each constant, and the constant matrix is ​​stored as an offline operation matrix for solving the Bezier curve. The preset Bezier curve equation is the Bezier curve equation used to describe the acceleration curve in the parking phase. The various unknown state parameters in the solution result matrix correspond to various parameters in the vehicle state vector. During the acceleration control process, the controller can solve the Bezier curve control points based on the vehicle state vector and the preset constant matrix to obtain the control point matrix of the actual Bezier curve, that is, the Bezier curve data is obtained by offline solving the Bezier curve by the Lagrange multiplier method. The control point matrix describes the control points of the Bezier curve corresponding to the current vehicle state.

[0102] In order to better illustrate the method provided by the embodiment of the present invention, the calculation process of the constant matrix is ​​briefly described here. Based on the description of the Bezier curve describing the acceleration of the parking phase in combination with equations (1) to (7) in the previous embodiment, on the basis of the Bezier curve describing the acceleration of the parking phase, refer to equations (8) to (16) as shown below to calculate the constraint matrix of the Bezier curve.

[0103] v2 m =v2 base (1), m∈[1,seg] (8)

[0104] V2=[v21 v22...v2 seg ] (9)

[0105] Cax0=[1 0 0...0]∈R 1* L, L = seg*(n+1) (10)

[0106] Cj0=[Jerk(0)0 0...0]∈R 1* L, L = seg*(n+1) (11)

[0107] Cax end =[0 0 0...1]∈R 1*L , L = seg*(n+1) (12)

[0108] C jend =[0 0 0...Jerk(1)]∈R 1*L , L = seg*(n+1) (13)

[0109]

[0110] First, through equations (8) and (9), the square increment of velocity v2 from the beginning to the end of a single Bezier curve can be calculated: m , and then calculate the total speed square increment V2 when each Bezier curve is connected together. In the actual acceleration control process, this value can be made equal to the calculated average deceleration to achieve the average deceleration of the entire planning curve.

[0111] The initial acceleration constraint is established by equation (10) to make the initial value of the Bezier curve equal to the acceleration at the control time point. The initial jerk constraint is established by equation (11) to make the initial slope of the Bezier curve equal to the jerk at the control time point. The terminal acceleration constraint is established by equation (12) to make the terminal value of the Bezier curve equal to the expected terminal acceleration. The terminal jerk constraint is established by equation (13) to make the terminal slope of the Bezier curve equal to the expected terminal jerk. The acceleration continuity constraint is established by equation (14) to make the Bezier curve segments continuous. The jerk continuity constraint is established by equation (15) to make the slopes of the Bezier curve segments continuous. Equation (16) represents the matrix after the matrices of the above constraints are combined, that is, the overall constraint matrix.

[0112] Referring to equations (17) to (23) shown below, the cost matrix can be calculated.

[0113]

[0114] H total =k jerk *H jerk +k snap *H snap (19)

[0115]

[0116] Specifically, by using equations (17) and (18), the sum of the squares of the overall jerk and the sum of the squares of the jerk of all curve segments can be calculated. m and DescendRatio2 m are vectors containing decreasing positive integers, which can be set according to actual needs. For example, they can be [2 1.71.4 1.1 0.8 0.5] and [2.2 1.6 1 0.7 0.350.14], which respectively represent that the cost of the jerk and the acceleration of the curve gradually decrease from the first section to the last section, which is used to balance the inconsistent time consumed by each section of the acceleration curve in the time coordinate system (the curve from the first section to the last section represents the same forward distance, but because the speed gradually decreases, the time consumed is actually from short to long). The cost function containing the calculation results of formula (17) and formula (18) can be calculated by formula (19), and the calculation result is the cost matrix, where k jerk Indicates the penalty coefficient for the Jerk (acceleration) of the curve, k snapIt represents the penalty coefficient for the snap (jerk) of the curve, for example, it can be set to 1. The larger the coefficient is, the lower the cost function value of the calculated curve is. Equations (20) and (21) represent the result matrices obtained by solving the results of equations (19) and (16) by the Lagrange multiplier method. The calculation results meet the constraints in the constraint matrix and minimize the cost in the cost matrix. By extracting constants from the result matrix, the matrices shown in equations (22) and (23) can be constructed. The matrices H2 and H3Seg const This is the matrix used for offline operations, where jerk end It represents the expected Jerk value at the end of the curve, which can be set according to actual needs, for example, it can be set to 0. const The constant matrix is ​​stored as a constant matrix in the acceleration control process to realize the offline calculation of the Bezier curve.

[0117] S203: Determine the auxiliary parameter t value corresponding to the current sampling time point;

[0118] In the method provided by an embodiment of the present invention, the controller can calculate the auxiliary parameter t value corresponding to the current sampling time point through parameters such as the current planned speed, advance acceleration, sampling period and the number of segments of the Bezier curve, that is, the t value corresponding to the current sampling time point in the Bezier curve t coordinate system.

[0119] S204: Solving the interpolation points of the Bezier curve according to the control point matrix and the auxiliary parameter t value, obtaining the acceleration corresponding to the auxiliary parameter t value, and using the acceleration corresponding to the auxiliary parameter t value as the curve planning acceleration.

[0120] In the method provided by an embodiment of the present invention, the controller uses a solution algorithm for the interpolation points of the Bezier curve. By controlling the point matrix and the t value corresponding to the current sampling time point, the controller can calculate the acceleration corresponding to the current sampling time point on the Bezier curve, and use the acceleration as the curve planning acceleration corresponding to the current sampling time point.

[0121] Based on the method provided in the embodiment of the present invention, the Bezier curve equation is solved in advance by the Lagrange multiplier method, and the constant matrix is ​​extracted from the solution result. During the acceleration control process, the Bezier curve can be planned and calculated offline based on the preset constant matrix. When the Bezier curve is solved in real time and the optimal calculation is achieved, the amount of calculation is small, the memory occupied by the controller is small, and it is conducive to improving the processing speed.

[0122] exist Figure 1On the basis of the method shown, in the method provided by the embodiment of the present invention, the process of performing slope limiting processing on the curve planning acceleration mentioned in step S107 to obtain the target acceleration corresponding to the curve planning acceleration includes:

[0123] performing a difference operation on the leading acceleration and the expected terminal acceleration, and taking the difference between the leading acceleration and the expected terminal acceleration as a target error;

[0124] In the method provided by the embodiment of the present invention, the controller may make a difference between the leading acceleration corresponding to the current sampling time point and the expected terminal acceleration, and use the difference between the two as the target error.

[0125] Based on the planned distance, the planned speed, the expected terminal acceleration, the leading acceleration and the preset expected stopping distance, a stopping distance error is estimated to obtain a stopping distance error corresponding to the current sampling time point;

[0126] In the method provided by the embodiment of the present invention, the controller can calculate the error between the distance moved when decelerating from the current acceleration to the expected terminal acceleration at a fixed jerk and the driving distance obtained by subtracting the expected stopping distance from the current distance, i.e., the stopping distance error, based on the current planned distance, planned speed, expected terminal acceleration, pre-acceleration, and the preset expected stopping distance. The preset expected stopping distance refers to the distance between the expected parking position of the vehicle and the obstacle ahead.

[0127] In a preset first slope set, a target acceleration slope soft upper limit value and a target acceleration slope soft lower limit value corresponding to the target error are determined; the first slope set includes a plurality of preset acceleration slope soft upper limit values ​​and a plurality of acceleration slope soft lower limit values;

[0128] In the method provided by the embodiment of the present invention, multiple acceleration slope soft upper limits and multiple acceleration slope soft lower limits are pre-set, and multiple acceleration errors or acceleration error ranges are pre-set, and each acceleration slope soft upper limit and each acceleration slope soft lower limit are set corresponding to each pre-set acceleration error or acceleration error range, that is, each acceleration slope soft upper limit and each acceleration slope soft lower limit have a corresponding acceleration error or acceleration error range. The first slope set is composed of each acceleration slope soft upper limit and each acceleration slope soft lower limit.

[0129] In the method provided by an embodiment of the present invention, during the acceleration control process, the controller can find the acceleration slope soft upper limit value and the acceleration slope soft lower limit value corresponding to the target error in the first slope set based on the target error, and use the found acceleration slope soft upper limit value as the target acceleration slope soft upper limit value, and use the found acceleration slope soft lower limit value as the target acceleration slope soft lower limit value.

[0130] Based on the planned speed and the parking distance error, determining a target acceleration slope upper limit value and a target acceleration slope lower limit value in a preset second slope set; the second slope set includes a plurality of preset acceleration slope upper limit values ​​and a plurality of acceleration slope lower limit values;

[0131] In the method provided by the embodiment of the present invention, a plurality of acceleration slope upper limits and acceleration slope lower limits are preset, and each acceleration slope upper limit and each acceleration slope lower limit are stored based on a certain correlation with the vehicle speed and the distance error, that is, a matching acceleration slope upper limit and acceleration slope lower limit can be found according to the specified vehicle speed and distance error. The second slope set is composed of each acceleration slope upper limit and acceleration slope lower limit.

[0132] During the acceleration control process, the controller can find the matching acceleration slope upper limit value and acceleration slope lower limit value in the second slope set according to the planned speed and parking distance error corresponding to the current sampling time point, thereby obtaining the target acceleration slope upper limit value and target acceleration slope lower limit value.

[0133] Determine the slope limit increment corresponding to the curve planning acceleration according to the target acceleration slope soft upper limit value, the target acceleration slope soft lower limit value, the target acceleration upper limit value and the target acceleration slope lower limit value;

[0134] In the method provided by an embodiment of the present invention, the controller determines the final upper slope boundary and lower slope boundary according to the target acceleration slope soft upper limit value, the target acceleration slope soft lower limit value, the target acceleration upper limit value and the target acceleration slope lower limit value, and then calculates the maximum slope increment that can be generated by the acceleration under the condition of slope limitation, i.e., the slope limit increment, based on the curve planning acceleration, the slope upper boundary and the slope lower boundary.

[0135] A sum operation is performed on the pre-acceleration and the slope limit increment, and the sum of the pre-acceleration and the slope limit increment is used as the target acceleration.

[0136] In the method provided by the embodiment of the present invention, the controller adds a slope limit increment to the preceding acceleration corresponding to the current sampling time point, and uses the calculated result as the current target acceleration.

[0137] Based on the method provided in the embodiment of the present invention, on the basis of obtaining the curve planning acceleration based on the Bezier curve, the slope of the final requested acceleration is further limited by combining parameters such as vehicle speed, acceleration error, and parking distance error, so that the slope of the acceleration change can be smaller, further ensuring the smoothness of the acceleration change.

[0138] In order to better illustrate the method provided by the embodiment of the present invention, based on the methods provided by the above embodiments and combined with actual application scenarios, the embodiment of the present invention provides another acceleration control method in the parking phase. The method provided by the embodiment of the present invention is applied to the scenario of vehicle following parking. The implementation process of the method provided by the embodiment of the present invention can be as follows: Figure 5 shown.

[0139] First, the offline operation matrix, i.e., the constant matrix mentioned in the previous embodiment, is calculated by the Bezier curve and the Lagrange multiplier method. The offline operation matrix is ​​stored in the controller for performing offline operation of the Bezier curve in the actual acceleration control process. For the description of the Bezier curve and the offline operation matrix describing the acceleration in the parking phase, please refer to the description provided in the previous embodiment in combination with equations (1) to (23), which will not be repeated here.

[0140] In order to fit the deceleration parking scenario, in the method provided in the embodiment of the present invention, the acceleration in this stage is represented by deceleration. The acceleration control process provided in the embodiment of the present invention mainly includes:

[0141] The controller can obtain relevant information about the vehicle in front and the vehicle itself through sensors. After the vehicle enters the parking phase, the controller obtains the vehicle speed vx (i.e., the vehicle speed in the previous embodiment) and the longitudinal distance sx from the vehicle in front (i.e., the obstacle distance in the previous embodiment) at each sampling time point, and obtains the requested deceleration ax at the previous sampling time point. lastReq (i.e., the leading acceleration at the current sampling time point in the previous embodiment). The unit of vx is (m / s), sx is specifically the longitudinal distance from the front bumper of the host vehicle to the center point of the rear of the target vehicle, and its unit is (m), ax lastReq The unit is (m / s 2 ).

[0142] The controller compensates for the vehicle speed and the longitudinal distance of the preceding vehicle by estimating the actuator lag. Specifically, since the vehicle actuator has a certain delay in responding to the requested deceleration, this embodiment calculates the vehicle speed vt (i.e., the planned vehicle speed in the previous embodiment) and the distance st (i.e., the planned distance in the previous embodiment) of the preceding vehicle after the expected actuator lag by using equations (24) and (25).

[0143]

[0144] Where delay is the estimated actuator delay time, for example, 0.02s, and a0 is the requested acceleration of the vehicle before delay seconds. Since the calculated result of vt may be a negative value, but the speed of the actual vehicle is generally not a negative number, st is calculated in different ways according to different vt. Lag compensation is helpful to reduce the control error caused by actuator lag and improve control stability.

[0145] The controller calculates the Jerk (i.e. the planned jerk mentioned above) and the t value (i.e. the auxiliary parameter t value mentioned above) of the vehicle in the t coordinate system. The Jerk value is expressed as t Indicates that the t value is expressed as t current The controller is calculated based on equation (26) to obtain t current , which is the value of t corresponding to the current requested acceleration in the first Bezier curve used in real-time planning. Based on formula (27), the Jerk in the t coordinate system can be calculated t value.

[0146]

[0147] Wherein, sampleTime represents the sampling period, which can be set to 0.02s, for example, Jerk lastReq ax lastReq The corresponding jerk. seg is the number of segments of the Bezier curve. da ,ax curretReq and ax raw The relevant calculations will be described later. It should be noted that when the current sampling time point is the first sampling time point when the vehicle enters the parking phase, the Jerk is calculated by the formula corresponding to the first operation in formula (27): t After the calculation process described below, the Jerk value at the current sampling time point will be obtained. da ,ax curretReq and ax raw And other parameters, and then according to the above parameters and ax at the current sampling time point lastreq , through the subsequent operation corresponding to the formula in formula (27), a Jerk can be calculated t , this Jerk t Then the Jerk at the next sampling time point t That is, the Jerk value applied in the acceleration planning process of the next sampling time point is calculated by using the parameters associated with each sampling time point and the formula corresponding to the subsequent operation in formula (27). t value.

[0148] The controller calculates the appropriate tail-end deceleration ae (i.e., the expected terminal acceleration in the previous embodiment) of the vehicle before parking through formula (28). This value is calculated only at the first sampling time point when the vehicle enters the parking stage, and each subsequent sampling time point uses the value calculated at the first sampling time point.

[0149]

[0150] Among them, tgtspeed is the preset terminal speed of the parking phase, axuplim is the preset upper limit of the vehicle tail end deceleration within the comfortable range, and axbotlim is the preset upper limit of the vehicle tail end deceleration within the comfortable range, for example, they are set to -0.4m / s 2 and -0.6m / s 2 The tail end deceleration obtained in this way can make the jerk of the overall acceleration curve smaller. The controller calculates the average deceleration range required to follow the target within the distance threshold range and the average deceleration used in the current plan through equations (29) to (32).

[0151]

[0152] Specifically, the upper limit of the average deceleration range is calculated by formula (29): That is, the average deceleration required to follow the vehicle to the position with the closest allowable following distance error. The lower limit of the average deceleration range is calculated by formula (30): That is, the average deceleration required when following and stopping at the position with the maximum allowable following and stopping distance error. The average deceleration currently applied can be calculated by formula (31): That is, the average deceleration sampled during acceleration planning at the previous sampling time point. If the current sampling time point is the first sampling time point, it is calculated according to the formula corresponding to the first operation in formula (31). If the current sampling time point is not the first sampling time point, the average deceleration calculated at the previous sampling time point is obtained. As The average deceleration for acceleration planning at the current sampling time point can be calculated by formula (32): (i.e. the expected average acceleration mentioned above), this formula indicates that the calculation starts from the average deceleration required for the zero stop error, gradually converges to ae, and satisfies the average deceleration range requirements. The average deceleration obtained based on this calculation method is conducive to achieving acceleration control with low noise and small vibration, so that the acceleration change converges smoothly to the expected tail deceleration within the allowable range of the stop distance error. Among them, tgtdist is the preset end distance of the parking stage, and timegain is the preset minimum average deceleration The time to converge to the desired tail end deceleration ae, for example, can be set to 1s.ErrThresUp and dx ErrThresBot They are respectively the preset upper and lower limits of the maximum allowable following distance error.

[0153] The controller uses formula (33) to calculate the following stop distance error S caused by maintaining the current deceleration to the tail end deceleration err (i.e., the parking distance error in the previous embodiment), that is, the moving distance when decelerating from the current deceleration at a fixed jerk to the tail end deceleration, and the error after subtracting the expected parking distance stopDist from the current distance, which is used to limit the slope of the requested deceleration.

[0154]

[0155] The controller calculates the requested deceleration ax based on the Bezier curve planning at the current sampling time point through equations (34) to (36): raw (i.e., the curve planning acceleration in the previous embodiment), that is, the Bessel planning deceleration curve and the requested deceleration at the current moment are calculated by the Lagrange multiplier method. At the same time, the controller can calculate the Jerk based on the Bessel curve planning at the current sampling time point through formula (37): da , which is used to calculate the Jerk for acceleration planning at the next sampling time point through formula (27) t .

[0156]

[0157] ax raw =AxCurr n (t current ) (36)

[0158]

[0159] Based on formula (34), the control point matrix AxCurr0 of the current Bezier curve can be calculated, where H2 and H3seg const The constant matrix extracted and constructed to solve the Bezier curve based on the Lagrange multiplier method can be referred to the description in the previous embodiment. The specific deceleration curve AxCurr can be drawn based on the control point matrix through formula (35): j , solve the acceleration value ax on the Bezier curve at the current sampling time point through equation (36) raw , that is, the requested deceleration based on Bezier curve planning.

[0160] After obtaining the requested deceleration based on Bezier curve planning, the controller limits the slope and amplitude of the requested deceleration through the following stop distance error, the current requested deceleration value and other information to obtain the final requested deceleration, and sends the final requested deceleration to the actuator to achieve acceleration control during the parking phase. Specifically, each acceleration slope soft upper limit, acceleration slope soft lower limit, acceleration slope upper limit and acceleration slope lower limit can be pre-set. In the acceleration planning process at the current sampling time point, the difference between the requested deceleration at the previous sampling time point and ae is used as the target error a Err , based on the target error table lookup, obtain the acceleration slope soft upper limit JerkTopLimitSoft (i.e. the target acceleration slope soft upper limit mentioned in the previous embodiment) and acceleration slope soft lower limit JerkBotLimitSoft (i.e. the target acceleration slope soft lower limit mentioned in the previous embodiment) corresponding to the target error. In the slope limiting strategy provided in the embodiment of the present invention, the greater the requested deceleration is than ae, the greater the absolute value of the positive value JerkTopLimitSoft is; and the greater the requested deceleration is than ae, the greater the absolute value of the negative value JerkBotLimitSoft is. The controller is based on the current vehicle speed vt and the following distance error S err Look up the table to obtain the acceleration slope upper limit JerkTopLimit (i.e., the target acceleration slope upper limit in the previous embodiment) and the acceleration slope lower limit JerkBotLimit (i.e., the target acceleration slope lower limit in the previous embodiment) that match the vehicle speed and the following stop distance error. err The larger the value (i.e. the car will stop too far), the larger the absolute value of JerkTopLimit is. err The smaller it is (i.e. it will stop too close), the larger the absolute value of the negative JerkTopLimit is. The requested acceleration ax calculated based on the Bezier curve is calculated by JerkTopLimitSoft, JerkBotLimitSoft, JerkTopLimit and JerkBotLimit raw Slope limit is performed. The maximum value of JerkTopLimitSoft and JerkTopLimit is taken as the maximum slope of the rise. If there is a positive jerk within 1s, the maximum slope of the descent is JerkBotLimit, otherwise it is the minimum value of JerkBotLimit and JerkBotLimitSoft. After the slope is limited by the maximum slope of the rise and the maximum slope of the descent, the calculated result is taken as the current final requested acceleration ax currentReq If the current maximum slope of the upward movement is JerkTopLimit, and the maximum slope of the downward movement is JerkBotLimit, ax currentReq The calculation formula can be shown as follows:

[0161] ax currentReq =ax lastReq +max(min((ax raw -ax lastReq ),JerkBotLimit*sampleTime),JerkTopLimit*sampleTime) (38)

[0162] Based on the method provided by the embodiment of the present invention, the parking plan is planned by the Lagrange multiplier method and the Bezier curve, so that the acceleration change in the parking stage is smooth. The calculation of the Bezier curve in real time is realized by the offline operation matrix, which can reduce the amount of calculation when the controller is running and save memory. The actuator lag compensation is used to reduce the following parking distance error caused by the actuator delay, and the actuator is easy to follow. On the basis of Bezier curve planning, combined with a series of range restrictions, an acceleration control scheme with small nodding, stable control, small setbacks, stable following distance, easy actuator following, small amount of calculation and memory usage can be achieved.

[0163] and Figure 1 Corresponding to the acceleration control method in the parking phase shown in FIG. 1 , an embodiment of the present invention further provides an acceleration control device in the parking phase, which is used to Figure 1 The specific implementation of the method shown in is shown in the structural diagram Figure 6 As shown, including:

[0164] The acquisition unit 301 is used to acquire the pre-acceleration, vehicle speed and obstacle distance corresponding to the current sampling time point when the controlled vehicle is in the parking stage; the pre-acceleration is the requested acceleration corresponding to the previous sampling time point of the current sampling time point, and the obstacle distance is the current distance between the controlled vehicle and the obstacle;

[0165] A first determining unit 302 is used to determine the planned speed and planned distance corresponding to the current sampling time point based on the vehicle speed and the obstacle distance;

[0166] A second determination unit 303 is used to determine the desired terminal acceleration corresponding to the controlled vehicle;

[0167] A third determining unit 304 is used to determine the expected average acceleration corresponding to the current sampling time point according to the planned speed, the planned distance, the leading acceleration and the expected terminal acceleration;

[0168] The fourth determining unit 305 is used to determine the planned acceleration corresponding to the current sampling time point;

[0169] a fifth determining unit 306, configured to determine Bezier curve data corresponding to the current sampling time point according to the expected average acceleration, the leading acceleration, the planned jerk, and the expected terminal acceleration, and determine the curve planned acceleration corresponding to the current sampling time point based on the Bezier curve data; the Bezier curve data represents a Bezier curve describing the acceleration of the vehicle during the parking phase;

[0170] The slope limiting unit 307 is used to perform slope limiting processing on the curve planning acceleration to obtain a target acceleration corresponding to the curve planning acceleration, and use the target acceleration as the requested acceleration corresponding to the current sampling time point to control the acceleration of the controlled vehicle.

[0171] exist Figure 6 Based on the device shown, the device provided in the embodiment of the present invention can be further expanded into multiple units. The function of each unit can be found in the description of each embodiment provided in the previous text for the acceleration control method in the parking phase, and no further examples will be given here.

[0172] An embodiment of the present invention further provides a storage medium, which includes stored instructions, wherein when the instructions are executed, the device where the storage medium is located is controlled to execute the acceleration control method in the parking phase as described above.

[0173] The embodiment of the present invention further provides an electronic device, the structural diagram of which is shown in FIG. Figure 7 As shown, it specifically includes a memory 401 and one or more instructions 402, wherein the one or more instructions 402 are stored in the memory 401 and are configured to be executed by one or more processors 403 to perform the following operations:

[0174] When the controlled vehicle is in the parking stage, the lead acceleration, vehicle speed and obstacle distance corresponding to the current sampling time point are obtained; the lead acceleration is the requested acceleration corresponding to the previous sampling time point of the current sampling time point, and the obstacle distance is the current distance between the controlled vehicle and the obstacle; based on the vehicle speed and the obstacle distance, the planned speed and planned distance corresponding to the current sampling time point are determined; the expected terminal acceleration corresponding to the controlled vehicle is determined; based on the planned speed, the planned distance, the lead acceleration and the expected terminal acceleration, the expected average acceleration corresponding to the current sampling time point is determined; the current the planned acceleration corresponding to the sampling time point; determining the Bezier curve data corresponding to the current sampling time point according to the expected average acceleration, the leading acceleration, the planned acceleration and the expected terminal acceleration, and determining the curve planned acceleration corresponding to the current sampling time point based on the Bezier curve data; the Bezier curve data represents a Bezier curve describing the vehicle acceleration during the parking phase; performing slope limiting processing on the curve planned acceleration to obtain a target acceleration corresponding to the curve planned acceleration, and using the target acceleration as the requested acceleration corresponding to the current sampling time point to control the acceleration of the controlled vehicle.

[0175] Each embodiment in this specification is described in a progressive manner, and the same and similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system or system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment. The system and system embodiment described above are only schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without creative work. Professionals can also further realize that the units and algorithm steps of each example described in combination with the embodiments disclosed in this article can be implemented in electronic hardware, computer software or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in the above description according to the function. Whether these functions are executed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present invention. The above description of the disclosed embodiments enables professional and technical personnel in the field to implement or use the present invention. Various modifications to these embodiments will be apparent to professional and technical personnel in the field, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but will conform to the widest range consistent with the principles and novel features disclosed herein.

Claims

1. A method for controlling acceleration during parking, characterized in that: include: When the controlled vehicle is in the parking stage, the pre-acceleration, vehicle speed and obstacle distance corresponding to the current sampling time point are obtained; the pre-acceleration is the requested acceleration corresponding to the previous sampling time point of the current sampling time point, and the obstacle distance is the current distance between the controlled vehicle and the obstacle; Determine the planned speed and planned distance corresponding to the current sampling time point based on the vehicle speed and the obstacle distance; Determining a desired terminal acceleration corresponding to the controlled vehicle; Determining an expected average acceleration corresponding to the current sampling time point according to the planned speed, the planned distance, the leading acceleration and the expected terminal acceleration; Determine the planned acceleration corresponding to the current sampling time point; Determine the Bezier curve data corresponding to the current sampling time point according to the expected average acceleration, the leading acceleration, the planned jerk and the expected terminal acceleration, and determine the curve planned acceleration corresponding to the current sampling time point based on the Bezier curve data; the Bezier curve data represents a Bezier curve describing the acceleration of the vehicle during the parking phase; The curve planning acceleration is subjected to slope limiting processing to obtain a target acceleration corresponding to the curve planning acceleration, and the target acceleration is used as the requested acceleration corresponding to the current sampling time point to control the acceleration of the controlled vehicle.

2. The method according to claim 1, characterized in that The determining, based on the vehicle speed and the obstacle distance, the planned speed and the planned distance corresponding to the current sampling time point includes: Performing speed time-delay compensation processing on the vehicle speed to obtain a time-delay estimated speed corresponding to the vehicle speed, and using the time-delay estimated speed as the planned speed; A distance time delay compensation process is performed on the obstacle distance to obtain a time delay estimated distance corresponding to the obstacle distance, and the time delay estimated distance is used as the planned distance.

3. The method according to claim 1, characterized in that The determining of the expected terminal acceleration corresponding to the controlled vehicle includes: If the current sampling time point is the first sampling time point when the controlled vehicle enters the parking phase, the estimated terminal acceleration is determined according to the preset parking phase terminal speed, the planned speed, the planned distance and the pre-acceleration; If the estimated terminal acceleration is within a preset terminal acceleration range, the estimated terminal acceleration is used as the optimal terminal acceleration corresponding to the current sampling time point; If the estimated terminal acceleration is not within the terminal acceleration range, determining the optimal terminal acceleration corresponding to the current sampling time point within the range threshold of the terminal acceleration range; The optimal terminal acceleration corresponding to the current sampling time point is stored, and the optimal terminal acceleration corresponding to the current sampling time point is used as the expected terminal acceleration; If the current sampling time point is not the first sampling time point, the stored optimal terminal acceleration corresponding to the first sampling time point is obtained, and the optimal terminal acceleration corresponding to the first sampling time point is used as the expected terminal acceleration.

4. The method according to claim 1, characterized in that: The determining, based on the planned speed, the planned distance, the leading acceleration and the expected terminal acceleration, the expected average acceleration corresponding to the current sampling time point includes: Determining an average acceleration range corresponding to the current sampling time point according to the planned speed, the planned distance, a preset parking phase end speed, a preset parking phase end distance, and a preset parking distance error range; Determine the current planned average acceleration corresponding to the current sampling time point; Determining an estimated average acceleration corresponding to the current sampling time point according to the expected terminal acceleration, the current planned average acceleration, the planned speed, the preceding acceleration and the terminal speed of the parking phase; If the estimated average acceleration is within the average acceleration range, the estimated average acceleration is used as the expected average acceleration.

5. The method according to claim 1, characterized in that The determining the planned acceleration corresponding to the current sampling time point includes: If the current sampling time point is the first sampling time point when the controlled vehicle enters the parking phase, determining the jerk corresponding to the pre-acceleration; Determine an estimated jerk corresponding to the current sampling time point based on preset Bezier curve parameters, the planned speed, the planned distance, the pre-acceleration, and the jerk corresponding to the pre-acceleration, and use the estimated jerk as the planned jerk; If the current sampling time point is not the first sampling time point, the curve planning jerk corresponding to the first sampling time point is determined, and the curve planning jerk is used as the planning jerk; the first sampling time point is a sampling time point before the current sampling time point, and the curve planning jerk is an acceleration determined based on the Bezier curve data corresponding to the first sampling time point, the request acceleration corresponding to the first sampling time point, and the pre-acceleration corresponding to the first sampling time point.

6. The method according to claim 1, characterized in that The determining of the Bezier curve data corresponding to the current sampling time point according to the expected average acceleration, the leading acceleration, the planned jerk and the expected terminal acceleration, and determining the curve planned acceleration corresponding to the current sampling time point based on the Bezier curve data, includes: Constructing a vehicle state vector based on the expected average acceleration, the leading acceleration, the planned jerk, and the expected terminal acceleration; Solving the Bezier curve control points according to the vehicle state vector and a preset constant matrix to obtain a control point matrix corresponding to the current sampling time point, and using the control point matrix as the Bezier curve data; the preset constant matrix is ​​a matrix constructed based on constants in a solution result matrix, and the solution result matrix is ​​a result obtained by solving a preset Bezier curve equation by a Lagrange multiplier method; Determine the auxiliary parameter t value corresponding to the current sampling time point; The Bezier curve interpolation points are solved according to the control point matrix and the auxiliary parameter t value to obtain the acceleration corresponding to the auxiliary parameter t value, and the acceleration corresponding to the auxiliary parameter t value is used as the curve planning acceleration.

7. The method according to claim 1, characterized in that The step of performing slope limiting processing on the curve planning acceleration to obtain a target acceleration corresponding to the curve planning acceleration includes: performing a difference operation on the leading acceleration and the expected terminal acceleration, and taking the difference between the leading acceleration and the expected terminal acceleration as a target error; Based on the planned distance, the planned speed, the expected terminal acceleration, the leading acceleration and the preset expected stopping distance, a stopping distance error is estimated to obtain a stopping distance error corresponding to the current sampling time point; In a preset first slope set, a target acceleration slope soft upper limit value and a target acceleration slope soft lower limit value corresponding to the target error are determined; the first slope set includes a plurality of preset acceleration slope soft upper limit values ​​and a plurality of acceleration slope soft lower limit values; Based on the planned speed and the parking distance error, determining a target acceleration slope upper limit value and a target acceleration slope lower limit value in a preset second slope set; the second slope set includes a plurality of preset acceleration slope upper limit values ​​and a plurality of acceleration slope lower limit values; Determine the slope limit increment corresponding to the curve planning acceleration according to the target acceleration slope soft upper limit value, the target acceleration slope soft lower limit value, the target acceleration upper limit value and the target acceleration slope lower limit value; A sum operation is performed on the pre-acceleration and the slope limit increment, and the sum of the pre-acceleration and the slope limit increment is used as the target acceleration.

8. An acceleration control device during parking phase, characterized in that: include: an acquisition unit, for acquiring, when the controlled vehicle is in a parking phase, a pre-acceleration, a vehicle speed, and an obstacle distance corresponding to a current sampling time point; the pre-acceleration is a requested acceleration corresponding to a sampling time point before the current sampling time point, and the obstacle distance is a distance between the controlled vehicle and the obstacle at present; A first determining unit, configured to determine a planned speed and a planned distance corresponding to the current sampling time point based on the vehicle speed and the obstacle distance; A second determination unit, used to determine the desired terminal acceleration corresponding to the controlled vehicle; A third determining unit, configured to determine an expected average acceleration corresponding to the current sampling time point according to the planned speed, the planned distance, the leading acceleration and the expected terminal acceleration; A fourth determining unit, configured to determine a planned jerk corresponding to the current sampling time point; a fifth determining unit, configured to determine Bezier curve data corresponding to the current sampling time point according to the expected average acceleration, the leading acceleration, the planned jerk, and the expected terminal acceleration, and determine the curve planned acceleration corresponding to the current sampling time point based on the Bezier curve data; the Bezier curve data represents a Bezier curve describing the acceleration of the vehicle during the parking phase; The slope limiting unit is used to perform slope limiting processing on the curve planning acceleration to obtain a target acceleration corresponding to the curve planning acceleration, and use the target acceleration as the requested acceleration corresponding to the current sampling time point to control the acceleration of the controlled vehicle.

9. A storage medium, characterized in that: The storage medium includes stored instructions, wherein when the instructions are executed, the device where the storage medium is located is controlled to execute the acceleration control method in the parking phase according to any one of claims 1 to 7.

10. An electronic device, characterized in that: The device comprises a memory and one or more instructions, wherein the one or more instructions are stored in the memory and configured to be executed by one or more processors to execute the acceleration control method in the parking phase as claimed in any one of claims 1 to 7.

Images