ACC Following Vehicle Target Acceleration Compensation Method and Module Based on Monocular Vision

Abstract

The present invention discloses a method and module for compensating the target acceleration of ACC following based on monocular vision. The method for compensating the target acceleration of ACC following based on monocular vision includes: acquiring the motion data of the host vehicle and the target vehicle; calculating the corrected acceleration value of the target vehicle; dividing the current scenario of the target vehicle into low-speed or high-speed scenarios, and adopting different weight coefficients for the acquired acceleration value of the target vehicle and the corrected acceleration value of the target vehicle; dividing the current working condition of the target vehicle into an acceleration working condition or a deceleration working condition; if the target vehicle is in a deceleration working condition or an acceleration working condition, querying a gain compensation value table, and calculating an acceleration gain compensation value according to the value obtained from the table and the target vehicle acceleration value after variable weighting; the acceleration gain compensation value is subjected to limiting and filtering processing to obtain the target acceleration value for following the vehicle required for calculating the expected acceleration of the host vehicle. The present invention can improve the control accuracy during vehicle following start and smooth vehicle following, and enhance the driving safety and comfort of its automatic control.

Description

Technical Field

[0001] The present invention relates to the field of automobiles, and particularly to a method for compensating the target acceleration of an ACC following vehicle based on monocular vision, and a module for compensating the target acceleration of an ACC following vehicle based on monocular vision. Background Art

[0002] With the increasing popularity of the Adaptive Cruise Control System (ACC), consumers have put forward higher challenges to the applicability of its functions. From the perspective of comfort, the ACC can automatically complete the smooth acceleration or deceleration of the vehicle, maintain a stable cruising speed or drive smoothly following the target vehicle ahead. From the perspective of safety, the ACC can handle the working conditions of a large braking deceleration of the vehicle ahead without the risk of collision. Nowadays, due to the influence of cost, the solution of the Advanced Driver Assistance System (ADAS) based on monocular vision occupies the mainstream position in the market. It not only has a lower cost, but also has a relatively good automatic driving effect for each auxiliary function. However, in some specific scenarios, especially when dealing with the working conditions of a large braking deceleration of the target vehicle, the vehicle itself often has a greater risk of collision. The reason is that there is an inaccurate estimation of the acceleration of the target vehicle identified by the monocular vision solution, especially in the working conditions of a large deceleration of the target vehicle. This results in an underestimation of the absolute value of the desired acceleration calculated by the vehicle itself, increasing the risk of collision with the vehicle ahead. At the same time, the control accuracy of starting to follow the vehicle and smoothly following the vehicle will also be affected to varying degrees.

[0003] In the Adaptive Cruise Control System (ACC), the automatic acceleration and deceleration control of following the target vehicle requires calculating the desired acceleration of the vehicle itself. Therefore, it is necessary to consider the motion states of the vehicle itself and the target vehicle in real time, such as the speed of the vehicle itself, the distance, speed, and acceleration of the target vehicle relative to the vehicle itself. However, at present, the ACC control method based on the monocular vision solution does not consider the inaccurate situation of identifying the acceleration of the target vehicle and introduces a method for compensating the target acceleration of following the vehicle, which is extremely likely to cause the risk of collision when following the vehicle, greatly reducing the applicability and safety of the system. Summary of the Invention

[0004] A series of simplified concepts are introduced in the Summary of the Invention section. These simplified concepts are all simplified from the prior art in this field, which will be further described in detail in the Detailed Description section. The Summary of the Invention section of the present invention does not mean to attempt to define the key features and essential technical features of the claimed technical solution, nor does it mean to attempt to determine the protection scope of the claimed technical solution.

[0005] The technical problem to be solved by the present invention is to provide a method that can compensate the target acceleration of following the vehicle for the working conditions where the acceleration of the target vehicle identified by monocular vision is inaccurate.

[0006] And, a module that can compensate the acceleration of the following vehicle target for the condition of inaccurate acceleration of the target vehicle in monocular vision recognition.

[0007] To solve the above technical problems, the ACC following vehicle target acceleration compensation method based on monocular vision provided by the present invention includes the following steps:

[0008] S1, Obtain the motion data of the host vehicle and the target vehicle;

[0009] S2, Calculate the acceleration correction value of the target vehicle according to the motion data of the host vehicle and the target vehicle, and obtain the acceleration correction value of the target vehicle;

[0010] S3, If the target vehicle is in a specified low-speed driving scenario, specify that the obtained acceleration value of the target vehicle has a relatively large weight;

[0011] If the target vehicle is in a specified high-speed driving scenario, specify that the acceleration correction value of the target vehicle has a relatively large weight;

[0012] The high- and low-speed driving scenarios can be divided according to performance and / or requirements. Exemplarily: when the speed of the target vehicle is lower than 8.33 m / s, it is a low-speed scenario. At this time, the weight value of the obtained acceleration value of the target vehicle is taken as 0.7, and the weight value of the acceleration correction value of the target vehicle is taken as 0.3. The weight value can be calibrated according to actual vehicle tests;

[0013] When the speed of the target vehicle is higher than 8.33 m / s, it is a high-speed scenario. At this time, the weight value of the obtained acceleration value of the target vehicle is taken as 0.2, and the weight value of the acceleration correction value of the target vehicle is taken as 0.8. The weight value can be calibrated according to actual vehicle tests;

[0014] According to the acceleration correction value of the target vehicle and the obtained acceleration value of the target vehicle and their respective weights, calculate the acceleration value of the target vehicle after variable weight, and then divide whether the target vehicle is currently in an acceleration condition or a deceleration condition;

[0015] Exemplarily, the calculation method of the acceleration value of the target vehicle is: the product of the obtained acceleration value of the target vehicle and its weight plus the product of the acceleration correction value of the target vehicle and its weight;

[0016] S4, If the target vehicle is in a deceleration condition, query the deceleration gain compensation value table with the acceleration value of the target vehicle after variable weight and the speed of the host vehicle, and multiply the table value by the acceleration correction value of the target vehicle to obtain the acceleration gain compensation value of the target vehicle under the deceleration condition;

[0017] Exemplarily, the deceleration gain compensation value table is a two-dimensional linear interpolation table. The rows represent acceleration values, and the columns represent speed values. The acceleration values are divided into -4, -3, -2, -1, -0.5, 0 m / s 2; The speed is divided into 1, 11.11, 19.44, 33.33 m / s. The gain compensation values corresponding to the rows and columns can be calibrated according to real vehicle tests, and this value is generally between 0.8 and 1.5;

[0018] If the target vehicle is in an acceleration condition, query the acceleration gain compensation value look-up table with the weighted target vehicle acceleration value and the host vehicle speed (this table is a two-dimensional linear interpolation table, the rows represent acceleration values, the columns represent speed values, and the acceleration values are divided into 0, 1, 2, 3 m / s 2 ; The speed is divided into 1, 11.11, 19.44, 33.33 m / s. The gain compensation values corresponding to the rows and columns can be calibrated according to real vehicle tests, and this value is generally between 0.8 and 1.5). Multiply the look-up table value by the target vehicle acceleration correction value to obtain the acceleration gain compensation value of the target vehicle under the acceleration condition;

[0019] Exemplarily, the acceleration gain compensation value look-up table is a two-dimensional linear interpolation table, the rows represent acceleration values, the columns represent speed values, and the acceleration values are divided into 0, 1, 2, 3 m / s 2 ; The speed is divided into 1, 11.11, 19.44, 33.33 m / s. The gain compensation values corresponding to the rows and columns can be calibrated according to real vehicle tests, and this value is generally between 0.8 and 1.5;

[0020] S5. The acceleration gain compensation value is processed by limiting and filtering to obtain the following vehicle target acceleration value required for calculating the host vehicle's expected acceleration.

[0021] Optionally, further improve the above-mentioned ACC following target acceleration compensation method based on monocular vision, including the motion data of the host vehicle and the target vehicle: the host vehicle speed, the target vehicle speed, the target vehicle acceleration, the target vehicle presence judgment flag bit, and the target vehicle ID update flag bit;

[0022] Optionally, further improve the above-mentioned ACC following target acceleration compensation method based on monocular vision, and calculate the target vehicle acceleration correction value through integral correction and filtering processing according to the motion data of the host vehicle and the target vehicle.

[0023] Optionally, further improve the above-mentioned ACC following target acceleration compensation method based on monocular vision. Integrate the target vehicle acceleration to obtain the uncorrected target vehicle speed. The initialization condition of the integration is the target vehicle ID update flag bit or the absence of the target vehicle ID;

[0024] Take the difference between the uncorrected target vehicle speed and the target vehicle speed, and use the product of the difference and the difference gain as the correction term of the target vehicle acceleration. The difference gain is obtained by two-dimensional look-up of the difference and the target vehicle speed;

[0025] The integral input term for the next operation cycle is the difference between the current target vehicle acceleration and the correction term of the target vehicle acceleration, and the target vehicle acceleration correction value is obtained through amplitude limiting and first-order filtering processing.

[0026] Optionally, further improve the ACC following target acceleration compensation method based on monocular vision, and the deceleration gain compensation value look-up table and the acceleration gain compensation value can be obtained through calibration.

[0027] To solve the above technical problems, the present invention provides an ACC following target acceleration compensation module based on monocular vision, including:

[0028] A state input unit that obtains the motion data of the host vehicle and the target vehicle from the host vehicle bus and the environment perception system;

[0029] A correction unit that calculates the target vehicle acceleration correction value according to the motion data of the host vehicle and the target vehicle, and obtains the target vehicle acceleration correction value;

[0030] A working condition division unit, if the target vehicle is in a specified low-speed driving scenario, specify that the obtained target vehicle acceleration value has a relatively large weight; if the target vehicle is in a specified high-speed driving scenario, specify that the target vehicle acceleration correction value has a relatively large weight;

[0031] According to the target vehicle acceleration correction value, the obtained target vehicle acceleration value and their respective weights, calculate the target vehicle acceleration value after variable weighting, and then divide whether the target vehicle is currently in an acceleration working condition or a deceleration working condition; a gain compensation unit, if the target vehicle is in a deceleration working condition, query the deceleration gain compensation value look-up table with the target vehicle acceleration value after variable weighting and the host vehicle speed, and multiply the look-up value by the target vehicle acceleration value after variable weighting to obtain the acceleration gain compensation value of the target vehicle under the deceleration working condition;

[0032] If the target vehicle is in an acceleration working condition, query the acceleration gain compensation value look-up table with the target vehicle acceleration value after variable weighting and the host vehicle speed, and multiply the look-up value by the target vehicle acceleration value after variable weighting to obtain the acceleration gain compensation value of the target vehicle under the acceleration working condition;

[0033] An output unit that obtains the following target acceleration value required for calculating the desired acceleration of the host vehicle through amplitude limiting and filtering processing of the acceleration gain compensation value.

[0034] Optionally, further improve the ACC following target acceleration compensation module based on monocular vision, and the motion data of the host vehicle and the target vehicle includes: the host vehicle speed, the target vehicle speed, the target vehicle acceleration, the target vehicle presence judgment flag bit, and the target vehicle ID update flag bit.

[0035] Optionally, further improve the ACC following target acceleration compensation module based on monocular vision. The correction unit calculates the target vehicle acceleration correction value through integral correction and filtering processing based on the motion data of the host vehicle and the target vehicle.

[0036] Optionally, further improve the ACC following target acceleration compensation module based on monocular vision. The integral correction and filtering processing performed by the correction unit includes:

[0037] Integrate the target vehicle acceleration to obtain the uncorrected target vehicle speed. The initialization condition for integration is the target vehicle ID update flag or the absence of the target vehicle ID.

[0038] Take the difference between the uncorrected target vehicle speed and the target vehicle speed, and use the product of the difference and the difference gain as the correction term for the target vehicle acceleration. The difference gain is obtained by two-dimensional lookup of the difference and the target vehicle speed.

[0039] The integral input term for the next operation cycle is the difference between the current target vehicle acceleration and the correction term for the target vehicle acceleration, and the target vehicle acceleration correction value is obtained through amplitude limiting and first-order filtering processing.

[0040] Optionally, further improve the ACC following target acceleration compensation module based on monocular vision. The deceleration gain compensation value lookup table and the acceleration gain compensation value can be obtained through calibration.

[0041] Considering the characteristic that the target vehicle acceleration identified by the monocular vision scheme is inaccurate, the present invention improves the accuracy of calculating the target vehicle acceleration, improves the control effect when dealing with large acceleration and deceleration following, and has practical significance for improving the control ability of the adaptive cruise control system (ACC). Its practical significance lies in that it realizes the extreme working conditions such as large braking deceleration of the ACC for the target vehicle, and at the same time can further improve the control accuracy during following start and smooth following, improve the driving safety and comfort of its automatic control, and has broad application prospects and high market value. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] The drawings of the present invention are intended to show the general characteristics of the methods, structures, and / or materials used in specific exemplary embodiments of the present invention to supplement the description in the specification. However, the drawings of the present invention are schematic diagrams not drawn to scale, and thus may not accurately reflect the precise structure or performance characteristics of any given embodiment. The drawings of the present invention should not be construed as limiting or restricting the scope of the numerical values or properties covered by the exemplary embodiments according to the present invention. The following further describes the present invention in detail with reference to the drawings and specific embodiments:

[0043] Figure 1 It is a schematic diagram of the framework of the present invention.

[0044] Figure 2 Schematic diagram of the principle of the acceleration calculation correction unit. Specific implementation manners

[0045] The following uses specific specific embodiments to illustrate the implementation manners of the present invention. Those skilled in the art can fully understand other advantages and technical effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through different specific implementation manners. The details in this specification can also be applied based on different viewpoints, and various modifications or changes can be made without departing from the overall design concept of the invention. It should be noted that, without conflict, the following embodiments and the features in the embodiments can be combined with each other. The following exemplary embodiments of the present invention can be implemented in many different forms and should not be construed as being limited only to the specific embodiments described herein. It should be understood that these embodiments are provided to make the disclosure of the present invention thorough and complete, and to fully convey the technical solutions of these exemplary specific embodiments to those skilled in the art.

[0046] The first embodiment;

[0047] The present invention provides a method for compensating the target acceleration of ACC following based on monocular vision, including the following steps:

[0048] S1, obtaining the motion data of the host vehicle and the target vehicle;

[0049] S2, calculating the acceleration correction value of the target vehicle according to the motion data of the host vehicle and the target vehicle, and obtaining the acceleration correction value of the target vehicle;

[0050] S3, if the target vehicle is in the specified low-speed driving scenario, specifying the obtained target vehicle acceleration value as a relatively large weight; S3, if the target vehicle is in the specified low-speed driving scenario, specifying the obtained target vehicle acceleration value as a relatively large weight;

[0051] If the target vehicle is in the specified high-speed driving scenario, specifying the acceleration correction value of the target vehicle as a relatively large weight;

[0052] According to the acceleration correction value of the target vehicle, the obtained target vehicle acceleration value and their respective weights, calculate the target vehicle acceleration value after variable weighting, and then divide whether the target vehicle is currently in an acceleration condition or a deceleration condition;

[0053] S4, if the target vehicle is in a deceleration condition, query the deceleration gain compensation value table with the target vehicle acceleration value after variable weighting and the host vehicle speed, and multiply the table value by the acceleration correction value of the target vehicle to obtain the acceleration gain compensation value of the target vehicle under the deceleration condition;

[0054] If the target vehicle is in an acceleration condition, query the acceleration gain compensation value look-up table with the target vehicle acceleration value after variable weighting and the host vehicle speed, and multiply the look-up value by the target vehicle acceleration correction value to obtain the acceleration gain compensation value of the target vehicle under the acceleration condition;

[0055] S5. The acceleration gain compensation value is processed by limiting and filtering to obtain the following vehicle target acceleration value required for calculating the expected acceleration of the host vehicle.

[0056] The second embodiment;

[0057] The present invention provides a following vehicle target acceleration compensation method based on monocular vision, including the following steps:

[0058] S1. Obtain the motion data of the host vehicle and the target vehicle; the motion data of the host vehicle and the target vehicle includes: the host vehicle speed, the target vehicle speed, the target vehicle acceleration, the target vehicle presence judgment flag bit, and the target vehicle ID update flag bit;

[0059] S2. Integrate the target vehicle acceleration to obtain the uncorrected target vehicle speed, and the initialization condition of the integration is the target vehicle ID update flag bit or the absence of the target vehicle ID;

[0060] Take the difference between the uncorrected target vehicle speed and the target vehicle speed, and use the product of the difference and the difference gain as the correction term of the target vehicle acceleration. The difference gain is obtained by two-dimensional look-up of the difference and the target vehicle speed;

[0061] The integration input term of the next operation cycle is the difference between the current target vehicle acceleration and the correction term of the target vehicle acceleration. After limiting amplitude and first-order filtering processing, the target vehicle acceleration correction value is obtained, and the target vehicle acceleration correction value is calculated;

[0062] S3. If the target vehicle is in a specified low-speed driving scenario, specify that the obtained target vehicle acceleration value has a relatively large weight; if the target vehicle is in a specified high-speed driving scenario, specify that the target vehicle acceleration correction value has a relatively large weight;

[0063] According to the target vehicle acceleration correction value, the obtained target vehicle acceleration value, and their respective weights, calculate the target vehicle acceleration value after variable weighting, and then divide whether the target vehicle is currently in an acceleration condition or a deceleration condition;

[0064] S4. If the target vehicle is in a deceleration condition, query the deceleration gain compensation value look-up table with the target vehicle acceleration value after variable weighting and the host vehicle speed, and multiply the look-up value by the target vehicle acceleration value after variable weighting to obtain the acceleration gain compensation value of the target vehicle under the deceleration condition; the deceleration gain compensation value look-up table and the acceleration gain compensation value can be obtained through calibration;

[0065] If the target vehicle is in an acceleration condition, query the acceleration gain compensation value look-up table with the weighted target vehicle acceleration value and the host vehicle speed, and multiply the look-up value by the weighted target vehicle acceleration value to obtain the acceleration gain compensation value for the target vehicle under the acceleration condition;

[0066] The third embodiment;

[0067] The present invention provides a monocular vision-based ACC following target acceleration compensation module, including:

[0068] A state input unit, which obtains the motion data of the host vehicle and the target vehicle from the host vehicle bus and the environment perception system;

[0069] A correction unit, which calculates the target vehicle acceleration correction value according to the motion data of the host vehicle and the target vehicle, and obtains the target vehicle acceleration correction value;

[0070] A working condition division unit, if the target vehicle is in a specified low-speed driving scenario, specify the obtained target vehicle acceleration value as a relatively large weight; if the target vehicle is in a specified high-speed driving scenario, specify the target vehicle acceleration correction value as a relatively large weight;

[0071] According to the target vehicle acceleration correction value, the obtained target vehicle acceleration value and their respective weights, calculate the weighted target vehicle acceleration value, and then divide whether the target vehicle is currently in an acceleration condition or a deceleration condition; a gain compensation unit, if the target vehicle is in a deceleration condition, query the deceleration gain compensation value look-up table with the weighted target vehicle acceleration value and the host vehicle speed, and multiply the look-up value by the weighted target vehicle acceleration value to obtain the acceleration gain compensation value for the target vehicle under the deceleration condition;

[0072] If the target vehicle is in an acceleration condition, query the acceleration gain compensation value look-up table with the weighted target vehicle acceleration value and the host vehicle speed, and multiply the look-up value by the weighted target vehicle acceleration value to obtain the acceleration gain compensation value for the target vehicle under the acceleration condition;

[0073] An output unit, which processes the acceleration gain compensation value through limiting and filtering to obtain the following target acceleration value required for calculating the desired acceleration of the host vehicle.

[0074] The fourth embodiment;

[0075] Reference Figure 1 Combined Figure 2 As shown, the present invention provides a monocular vision-based ACC following target acceleration compensation module, including:

[0076] A state input unit, which obtains the motion data of the host vehicle and the target vehicle from the host vehicle bus and the environment perception system;

[0077] Motion data of the host vehicle and the target vehicle, including: host vehicle speed, target vehicle speed, target vehicle acceleration, target vehicle presence judgment flag bit, and target vehicle ID update flag bit;

[0078] Considering the influence of the drive types (front-wheel drive, rear-wheel drive, four-wheel drive), road curvature, and ground adhesion conditions of different vehicle models, the host vehicle speed is calculated after processing the four-wheel wheel speeds. When the target vehicle ID (sent by the monocular vision scheme) exists (judged as having a following target), the calculated host vehicle speed and the relative speed of the target vehicle recognized by the monocular vision scheme with respect to the host vehicle are added to obtain the target vehicle speed VIVSpd. And when the target vehicle IDs sent in the front and rear frames are different, it is judged that the target vehicle ID is updated. The acceleration of the target vehicle recognized by the monocular vision scheme is used as the input value of the target vehicle acceleration for the acceleration calculation correction unit;

[0079] Correction unit, which calculates the correction value of the target vehicle acceleration according to the motion data of the host vehicle and the target vehicle, and obtains the correction value of the target vehicle acceleration;

[0080] Operating condition division unit, if the target vehicle is in a specified low-speed driving scenario, the specified obtained target vehicle acceleration value has a relatively large weight; if the target vehicle is in a specified high-speed driving scenario, the specified target vehicle acceleration correction value has a relatively large weight;

[0081] According to the target vehicle acceleration correction value, the obtained target vehicle acceleration value, and their respective weights, the target vehicle acceleration value after variable weighting is calculated, and then it is divided into whether the target vehicle is currently in an acceleration operating condition or a deceleration operating condition;

[0082] Gain compensation unit, if the target vehicle is in a deceleration operating condition, look up the deceleration gain compensation value table with the target vehicle acceleration value after variable weighting and the host vehicle speed, and multiply the looked-up value by the target vehicle acceleration value after variable weighting to obtain the acceleration gain compensation value under the deceleration operating condition of the target vehicle;

[0083] If the target vehicle is in an acceleration operating condition, look up the acceleration gain compensation value table with the target vehicle acceleration value after variable weighting and the host vehicle speed, and multiply the looked-up value by the target vehicle acceleration value after variable weighting to obtain the acceleration gain compensation value under the acceleration operating condition of the target vehicle;

[0084] Exemplarily, it is further described as follows;

[0085] Integrate the target vehicle acceleration input value a AcceltnInput to obtain the uncorrected target vehicle speed V EstimatedVehSpd , and the initialization condition of the integration is that a new target is detected (target vehicle ID update flag bit) or there is no target vehicle ID (no following target), and the initialization value is the target vehicle speed value V IVSpd . Then, V EstimatedVehSpdSubtract from the target vehicle speed V obtained from the previous calculation step IVSpd Take the difference, and the product of the difference and the difference gain is used as the correction term a for the target vehicle acceleration AcceltnErr , where the difference gain is obtained by two-dimensional look-up based on the difference and V IVSpd . The integral input term for the next operation cycle is the difference a between the target vehicle acceleration input value at this time and the correction term a for the target vehicle acceleration AcceltnErr . Thus, a AcceltnUnfiltered . After being limited in amplitude and processed by first-order filtering, a AcceltnUnfiltered is obtained AcceltnModified .

[0086] The following vehicle target acceleration correction value a processed by the acceleration calculation correction unit AcceltnModified is compared with the target vehicle acceleration input value a AcceltnInput . In the case of a low-speed scenario of the leading vehicle (when the target vehicle speed is lower than 8.33 m / s, it is a low-speed scenario), take a AcceltnInput as the larger weight (at this time, a AcceltnInput has a higher credibility, the weight value of a AcceltnInput is taken as 0.7, and the weight value of a AcceltnUnfiltered is taken as 0.3, and the weight value can be calibrated according to actual vehicle tests) to divide the acceleration and deceleration scenarios of the target vehicle; in the case of a high-speed scenario of the leading vehicle (when the target vehicle speed is higher than 8.33 m / s, it is a high-speed scenario), take a AcceltnModified as the larger weight (at this time, a AcceltnInput has a lower credibility, the weight value of a AcceltnInput is taken as 0.2, and a AcceltnUnfiltered is taken as 0.8, and the weight value can be calibrated according to actual vehicle tests) to divide the acceleration and deceleration scenarios of the target vehicle. The product of the obtained target vehicle acceleration value (a AcceltnInput ) and its weight plus the product of the target vehicle acceleration correction value (a AcceltnUnfiltered ) and its weight gives the target vehicle acceleration value a AcceltnWeighted after variable weighting. The scenarios are divided into the target vehicle deceleration working condition and the target vehicle acceleration working condition. For the target vehicle deceleration working condition, use the following vehicle target variable-weight acceleration value and the own vehicle speed as two-dimensional information inputs to look up the gain compensation value in a table, and the product of the look-up value and the following vehicle target variable-weight acceleration value gives the acceleration gain compensation value in the target vehicle deceleration working condition. Similarly, for the target vehicle acceleration working condition, use the following vehicle target variable-weight acceleration value and the own vehicle speed as two-dimensional information inputs to look up the gain compensation value in a table, and the product of the look-up value and the following vehicle target variable-weight acceleration value gives the acceleration gain compensation value in the target vehicle acceleration working condition. The above look-up value determines the gain value through actual vehicle tests according to the vehicle models and the accuracy of the leading vehicle acceleration recognized by the monocular vision scheme;

[0087] Output unit, which obtains the following vehicle target acceleration value required for calculating the expected acceleration of the own vehicle by limiting the amplitude and filtering the acceleration gain compensation value;

[0088] Among them, the clipping process is as follows: a limit is set for both the increase and decrease of the acceleration in each operation cycle to eliminate the abnormal acceleration fluctuations brought about during the processing of the acceleration gain compensation unit.

[0089] The filtering process is as follows: the acceleration after clipping is further subjected to a first-order filtering process, and the filtering coefficient value can be adjusted according to the actual vehicle test conditions to smooth the calculated acceleration value of the target vehicle.

[0090] Unless otherwise defined, all terms (including technical and scientific terms) used herein shall have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. It will also be understood that terms such as those defined in a general dictionary shall be interpreted as having a meaning consistent with their meaning in the context of the relevant art, and shall not be interpreted in an idealized or overly formal sense unless expressly defined herein.

[0091] The above has described the present invention in detail through specific implementation manners and examples, but these do not constitute a limitation to the present invention. Without departing from the principle of the present invention, those skilled in the art can also make many deformations and improvements, which should also be regarded as the protection scope of the present invention.

Claims

1. A method for compensating the target acceleration of ACC following vehicle based on monocular vision, characterized in that, It includes the following steps: S1. Obtain the motion data of the host vehicle and the target vehicle; S2. Calculate the acceleration correction value of the target vehicle based on the motion data of the host vehicle and the target vehicle, and obtain the acceleration correction value of the target vehicle; S3. If the target vehicle is in a specified low-speed driving scenario, specify that the obtained acceleration value of the target vehicle has a relatively large weight; If the target vehicle is in a specified high-speed driving scenario, specify that the acceleration correction value of the target vehicle has a relatively large weight; According to the acceleration correction value of the target vehicle, the obtained acceleration value of the target vehicle, and their respective weights, calculate the acceleration value of the target vehicle after variable weighting, and then determine whether the target vehicle is currently in an acceleration condition or a deceleration condition; S4. If the target vehicle is in a deceleration condition, query the deceleration gain compensation value look-up table with the acceleration value of the target vehicle after variable weighting and the host vehicle speed, and multiply the look-up value by the acceleration correction value of the target vehicle to obtain the acceleration gain compensation value of the target vehicle under the deceleration condition; If the target vehicle is in an acceleration condition, query the acceleration gain compensation value look-up table with the acceleration value of the target vehicle after variable weighting and the host vehicle speed, and multiply the look-up value by the acceleration correction value of the target vehicle to obtain the acceleration gain compensation value of the target vehicle under the acceleration condition; S5. The acceleration gain compensation value is subjected to amplitude limiting and filtering processing to obtain the following vehicle target acceleration value required for calculating the desired acceleration of the host vehicle.

2. The ACC following vehicle target acceleration compensation method based on monocular vision according to claim 1, characterized in that: The motion data of the host vehicle and the target vehicle includes: the host vehicle speed, the target vehicle speed, the target vehicle acceleration, the target vehicle presence judgment flag bit, and the target vehicle ID update flag bit.

3. The ACC following vehicle target acceleration compensation method based on monocular vision according to claim 1, characterized in that: The acceleration correction value of the target vehicle is calculated based on the motion data of the host vehicle and the target vehicle through integral correction and filtering processing.

4. The ACC following vehicle target acceleration compensation method based on monocular vision according to claim 3, characterized in that: Integrate the acceleration of the target vehicle to obtain the uncorrected target vehicle speed, and the initialization condition of the integration is the target vehicle ID update flag bit or the absence of the target vehicle ID; Take the difference between the uncorrected target vehicle speed and the target vehicle speed, and use the product of the difference and the difference gain as the correction term of the target vehicle acceleration. The difference gain is obtained by two-dimensional look-up of the difference and the target vehicle speed; The integration input term of the next operation cycle is the difference between the current target vehicle acceleration and the correction term of the target vehicle acceleration, and the acceleration correction value of the target vehicle is obtained through amplitude limiting and first-order filtering processing.

5. The method for compensating the target acceleration of ACC vehicle following based on monocular vision according to claim 1, characterized in that: The deceleration gain compensation value look-up table and the acceleration gain compensation value can be obtained through calibration.

6. A target acceleration compensation module for ACC following based on monocular vision, characterized in that, It includes: A state input unit that obtains the motion data of the host vehicle and the target vehicle from the host vehicle bus and the environment perception system; A correction unit that calculates the acceleration correction value of the target vehicle based on the motion data of the host vehicle and the target vehicle, and obtains the acceleration correction value of the target vehicle; The driving condition division unit assigns a relatively large weight to the target vehicle acceleration value obtained if the target vehicle is in a specified low-speed driving scenario, and assigns a relatively large weight to the target vehicle acceleration correction value if the target vehicle is in a specified high-speed driving scenario. Based on the target vehicle acceleration correction value, the obtained target vehicle acceleration value, and their respective weights, calculate the target vehicle acceleration value with variable weights, and then divide whether the target vehicle is currently in an acceleration driving condition or a deceleration driving condition. The gain compensation unit, if the target vehicle is in a deceleration driving condition, looks up the deceleration gain compensation value in a table using the target vehicle acceleration value with variable weights and the self-vehicle speed, and multiplies the looked-up value by the target vehicle acceleration value with variable weights to obtain the acceleration gain compensation value for the target vehicle in the deceleration driving condition. If the target vehicle is in an acceleration driving condition, looks up the acceleration gain compensation value in a table using the target vehicle acceleration value with variable weights and the self-vehicle speed, and multiplies the looked-up value by the target vehicle acceleration value with variable weights to obtain the acceleration gain compensation value for the target vehicle in the acceleration driving condition. The output unit processes the acceleration gain compensation value through amplitude limiting and filtering to obtain the following vehicle target acceleration value required for calculating the self-vehicle expected acceleration.

7. The monocular vision-based ACC following vehicle target acceleration compensation module according to claim 6, wherein: The motion data of the self-vehicle and the target vehicle includes: the self-vehicle speed, the target vehicle speed, the target vehicle acceleration, the target vehicle presence judgment flag bit, and the target vehicle ID update flag bit.

8. The monocular vision-based ACC following vehicle target acceleration compensation module according to claim 6, wherein: The correction unit calculates the target vehicle acceleration correction value through integral correction and filtering processing based on the motion data of the self-vehicle and the target vehicle.

9. The monocular vision-based ACC following vehicle target acceleration compensation module according to claim 8, wherein: The correction unit performs integral correction and filtering processing including: Integrating the target vehicle acceleration to obtain the uncorrected target vehicle speed, and the initialization condition of the integration is the target vehicle ID update flag bit or the absence of the target vehicle ID; Taking the difference between the uncorrected target vehicle speed and the target vehicle speed, and using the product of the difference and the difference gain as the correction term of the target vehicle acceleration, where the difference gain is obtained by two-dimensional table lookup of the difference and the target vehicle speed; The integral input term of the next operation cycle is the difference between the current target vehicle acceleration and the correction term of the target vehicle acceleration, and the target vehicle acceleration correction value is obtained through amplitude limiting and first-order filtering processing.

10. The ACC following vehicle target acceleration compensation module based on monocular vision according to claim 6, wherein: The deceleration gain compensation value table and the acceleration gain compensation value can be obtained through calibration.

Images