Filtering method, device and electronic equipment for intelligent driving perception system

Abstract

The present invention provides a filtering method, device and electronic device for an intelligent driving perception system. The method includes: determining a target motion model set of an obstacle target based on a motion model in a priori knowledge base, judging whether the target motion model set is consistent with the model set of a VSIMM filter at the previous moment; if not, determining the target model set of the VSIMM filter at the current moment based on the target motion model set and the model set of the VSIMM filter at the previous moment; obtaining measurement information of the obstacle target at the current moment, so that the VSIMM filter estimates the filtering state according to the measurement information and the target model set, and obtains the total estimated output information of the VSIMM filter at the current moment; tracking the obstacle target according to the total estimated output information, thereby improving the tracking accuracy of the obstacle target through the VSIMM filter, thereby improving the performance of the intelligent driving perception system, and having good practical value.

Description

Technical Field

[0001] The present invention relates to the field of intelligent driving environment perception technology, and in particular to a filtering method, device and electronic equipment for an intelligent driving perception system. Background Art

[0002] In the intelligent driving perception system, to improve its performance, it is mainly to improve the ability of obstacle detection and obstacle target tracking, among which target tracking is a very important link. The filter is used in the target tracking process to calculate the position, speed, trajectory, number, type and characteristics of obstacles in the surrounding environment. In the field of intelligent driving, commonly used sensors are generally lidar, camera, millimeter wave radar and ultrasonic radar.

[0003] There are two main computational optimization problems faced by target tracking filtering: first, the obstacles in the intelligent driving environment are highly maneuverable and vary greatly, making them difficult to predict; second, the vehicle environment is complex and contains a lot of noise, and the sensor itself also has certain error noise, which is difficult to fit, resulting in problems such as low filter state estimation accuracy and even filter divergence during target tracking. Existing methods mainly use adaptive filtering algorithms for filtering. Although this method can improve the accuracy of obstacle target tracking to a certain extent, due to the strong maneuverability and randomness of obstacle targets, it is difficult to determine the motion model of the obstacle target, thereby affecting the filtering effect, resulting in a decrease in the accuracy of obstacle target tracking, and even the loss of obstacle targets, affecting the performance of the intelligent driving perception system. Summary of the invention

[0004] In view of this, an object of the present invention is to provide a filtering method, device and electronic device for an intelligent driving perception system to alleviate the above-mentioned problems.

[0005] In a first aspect, an embodiment of the present invention provides a filtering method for an intelligent driving perception system, the method being applied to a server of the intelligent driving perception system, wherein the server is provided with a priori knowledge base and a variable structure interactive multi-model VSIMM filter, the method comprising: determining a target motion model set of an obstacle target based on a motion model in the priori knowledge base, wherein the motion model comprises at least one of the following: a uniform velocity CV model, a uniform acceleration CA model, a uniform rotation speed CT model, a current Current model, and a curved motion CM model; judging whether the target motion model set is consistent with a model set of the VSIMM filter at a previous moment; wherein the model set of the VSIMM filter at a previous moment comprises a motion model corresponding to each filter in the VSIMM filter at a previous moment; if not, determining a target model set of the VSIMM filter at a current moment based on the target motion model set and the model set of the VSIMM filter at a previous moment; obtaining measurement information of the obstacle target at a current moment, so that the VSIMM filter performs filtering state estimation according to the measurement information and the target model set, and obtains total estimated output information of the VSIMM filter at a current moment; and tracking the obstacle target according to the total estimated output information.

[0006] In combination with the first aspect, an embodiment of the present invention provides a first possible implementation method of the first aspect, wherein the step of determining whether the target motion model set is consistent with the model set of the VSIMM filter at the previous moment includes: determining whether each target motion model in the target motion model set is exactly the same as the motion model corresponding to each filter in the VSIMM filter at the previous moment; if so, determining that the target motion model set is consistent with the model set of the VSIMM filter at the previous moment; if there is any difference, determining that the target motion model set is inconsistent with the model set of the VSIMM filter at the previous moment.

[0007] In combination with the first possible implementation of the first aspect, an embodiment of the present invention provides a second possible implementation of the first aspect, wherein the step of determining the target model set of the VSIMM filter at the current moment based on the target motion model set and the model set of the VSIMM filter at the previous moment includes: obtaining the model set of the VSIMM filter at the current moment according to the union of the target motion model set and the model set of the VSIMM filter at the previous moment; obtaining a first model probability of the target motion model set and a second model probability of the model set of the VSIMM filter at the previous moment; and determining the target model set of the VSIMM filter at the current moment according to the ratio of the first model probability to the second model probability.

[0008] In combination with the second possible implementation of the first aspect, an embodiment of the present invention provides a third possible implementation of the first aspect, wherein the step of determining the target model set of the VSIMM filter at the current moment based on the ratio of the first model probability to the second model probability includes: judging whether the ratio is greater than a first threshold, and if so, determining that the target model set of the VSIMM filter at the current moment is a target motion model set; if not, judging whether the ratio is less than a second threshold, and if so, determining that the target model set of the VSIMM filter at the current moment is the model set of the VSIMM filter at the previous moment; wherein the second threshold is less than the first threshold.

[0009] In combination with the third possible implementation of the first aspect, an embodiment of the present invention provides a fourth possible implementation of the first aspect, wherein the step of estimating the filtering state of the VSIMM filter according to the measurement information and the target model set includes: determining the corresponding target filter in the VSIMM filter according to each target model in the target model set; performing filtering state estimation on each target filter according to the measurement information and the corresponding target model to obtain the total estimated output information of the VSIMM filter at the current moment; wherein the total estimated output information includes a total state estimation value and a total error covariance, the total state estimation value includes the state estimation value of each target filter at the current moment, and the total error covariance includes the error covariance of each target filter at the current moment.

[0010] In combination with the fourth possible implementation of the first aspect, an embodiment of the present invention provides a fifth possible implementation of the first aspect, wherein each target model is also configured with a model probability, and the method further includes: updating the model probability of each target model based on the likelihood function of each target model; wherein the likelihood function is calculated by the following formula:

[0011]

[0012] Among them, Λ j (k) represents the likelihood function of the jth target model at time k, S j (k) represents the measurement covariance of the jth target model at time k, v j (k) represents the Kalman filter residual of the j-th target model at time k.

[0013] In combination with the first aspect, an embodiment of the present invention provides a sixth possible implementation of the first aspect, wherein the step of determining a target motion model set of an obstacle target based on the motion model in a priori knowledge base includes: obtaining acquisition information of the obstacle target; wherein the acquisition information includes type information, location information and parameter information; determining a first motion model set of the obstacle target based on the type information; determining a second motion model set of the obstacle target based on the location information; determining a third motion model set of the obstacle target based on the parameter information; wherein the parameter information includes at least one of the following: acquisition device information, obstacle target profile information and meteorological information; determining a target motion model set of the obstacle target based on the first motion model set, the second motion model set and the third motion model set.

[0014] In a second aspect, an embodiment of the present invention further provides a filtering device of an intelligent driving perception system, which is applied to a server of the intelligent driving perception system, wherein the server is provided with a priori knowledge base and a variable structure interactive multi-model VSIMM filter, and the device comprises: a first determination module, for determining a target motion model set of an obstacle target based on a motion model in the priori knowledge base, wherein the motion model comprises at least one of the following: a uniform speed CV model, a uniform acceleration CA model, a uniform rotation speed CT model, a current Current model and a curved motion CM model; a judgment module, for judging whether the target motion model set is consistent with the model set of the VSIMM filter at the previous moment whether they are consistent; wherein the model set of the VSIMM filter at the previous moment includes the motion model corresponding to each filter in the VSIMM filter at the previous moment; a second determination module is used to determine the target model set of the VSIMM filter at the current moment based on the target motion model set and the model set of the VSIMM filter at the previous moment if not; a filter estimation module is used to obtain the measurement information of the obstacle target at the current moment, so that the VSIMM filter estimates the filtering state according to the measurement information and the target model set, and obtains the total estimated output information of the VSIMM filter at the current moment; a tracking module is used to track the obstacle target according to the total estimated output information.

[0015] In a third aspect, an embodiment of the present invention further provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the steps of the filtering method of the intelligent driving perception system of the first aspect mentioned above are implemented.

[0016] In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the filtering method of the intelligent driving perception system of the first aspect are executed.

[0017] The embodiments of the present invention bring the following beneficial effects:

[0018] The embodiments of the present invention provide a filtering method, device and electronic device for an intelligent driving perception system, which dynamically determines the target motion model of an obstacle target based on a motion model in a priori knowledge base, and determines the target model set of a VSIMM filter at the current moment according to the target motion model for filtering processing, so that the VSIMM filter can converge quickly when the obstacle target maneuvers, thereby improving the tracking accuracy of the obstacle target, thereby improving the performance of the intelligent driving perception system, and having good practical value.

[0019] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention are realized and obtained by the structures particularly pointed out in the description and the drawings.

[0020] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0022] Figure 1 A schematic diagram of a filtering method for an intelligent driving perception system provided by an embodiment of the present invention;

[0023] Figure 2 A flowchart of a filtering method for an intelligent driving perception system provided by an embodiment of the present invention;

[0024] Figure 3 A road terrain schematic diagram provided by an embodiment of the present invention;

[0025] Figure 4 A schematic diagram of a target motion model set for constructing an obstacle target provided by an embodiment of the present invention;

[0026] Figure 5 A schematic diagram of determining a target model set provided by an embodiment of the present invention;

[0027] Figure 6 A working principle diagram of a VSIMM filter provided by an embodiment of the present invention;

[0028] Figure 7 A schematic diagram of a filtering device for an intelligent driving perception system provided by an embodiment of the present invention;

[0029] Figure 8 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0030] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the described embodiments are 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.

[0031] In view of the problem that the motion model of existing obstacle targets is difficult to determine, thereby affecting the filtering effect, the embodiments of the present invention provide a filtering method, device and electronic equipment for an intelligent driving perception system, which improves the tracking accuracy of obstacle targets and further improves the performance of the intelligent driving perception system, and has good practical value.

[0032] To facilitate understanding of this embodiment, a filtering method for an intelligent driving perception system provided by an embodiment of the present invention is first introduced in detail below.

[0033] The embodiment of the present invention provides a filtering method for an intelligent driving perception system, which is applied to the intelligent driving perception system. The intelligent driving perception system includes a server and a detection unit connected to the server, wherein the detection unit includes multiple detection elements such as sensors, etc., so as to obtain environmental perception information of the intelligent driving perception system, wherein the environmental perception information includes sensor information, electronic map information, and other environmental information.

[0034] like Figure 1 As shown, the server constructs a priori knowledge base based on the acquired environmental perception information, and the priori knowledge base outputs the corresponding filter motion model set, and adjusts the model set of the VSIMM filter based on the model probability calculated by the filter motion model set and the VSIMM (Variable Structure Interacting Multiple Model) filter, and performs filter state estimation based on the adjusted VSIMM filter according to the obstacle motion data of the input obstacle target, so as to track the obstacle target. It should be noted that the above-mentioned VSIMM filter is also called VSIMM filtering algorithm or VSIMM algorithm.

[0035] Based on the above server, a filtering method of an intelligent driving perception system provided by an embodiment of the present invention is as follows Figure 2 As shown, the method comprises the following steps:

[0036] Step S202, determining a target motion model set of the obstacle target based on the motion model in the prior knowledge base;

[0037] Among them, in the intelligent driving perception system, the motion model of the obstacle target includes at least one of the following: CV (Constant Velocity, uniform speed) model, CA (Constant Acceleration, uniform acceleration) model, CT (Constant Turn, uniform speed) model, current model and CM (Curvilinear Motion, curvilinear motion) model. It should be noted that the motion models of other obstacle targets can be set according to actual conditions, and the embodiments of the present invention do not limit this description.

[0038] For ease of understanding, each motion model is described as follows:

[0039] (1) CV model; wherein the CV model is calculated according to the following formula:

[0040]

[0041] Among them, x k represents the displacement of the obstacle target in the x direction at time k, represents the speed of the obstacle target in the x direction at time k, y k represents the displacement of the obstacle target in the y direction at time k, represents the speed of the obstacle target in the y direction at time k, x k+1 represents the displacement of the obstacle target in the x direction at time k+1, represents the speed of the obstacle target in the x direction at time k+1, y k+1 represents the displacement of the obstacle target in the y direction at time k+1, represents the speed of the obstacle target in the y direction at time k+1, and T represents the time interval, where the time interval is the time interval from time k to time k+1.

[0042] (2) CA model; wherein the CA model is calculated according to the following formula:

[0043]

[0044] Among them, x k represents the displacement of the obstacle target in the x direction at time k, represents the speed of the obstacle target in the x direction at time k, represents the acceleration of the obstacle target in the x direction at time k, y k represents the displacement of the obstacle target in the y direction at time k, represents the speed of the obstacle target in the y direction at time k, represents the acceleration of the obstacle target in the y direction at time k, x k+1 represents the displacement of the obstacle target in the x direction at time k+1, represents the speed of the obstacle target in the x direction at time k+1, represents the acceleration of the obstacle target in the x direction at time k+1, y k+1 represents the displacement of the obstacle target in the y direction at time k+1, represents the speed of the obstacle target in the y direction at time k+1, represents the acceleration of the obstacle target in the y direction at time k+1, and T represents the time interval.

[0045] (3) CT model; wherein the CT model is calculated according to the following formula:

[0046]

[0047] Among them, x k represents the displacement of the obstacle target in the x direction at time k, represents the speed of the obstacle target in the x direction at time k, y k represents the displacement of the obstacle target in the y direction at time k, represents the speed of the obstacle target in the y direction at time k, x k+1 represents the displacement of the obstacle target in the x direction at time k+1, represents the speed of the obstacle target in the x direction at time k+1, y k+1 represents the displacement of the obstacle target in the y direction at time k+1, represents the speed of the obstacle target in the y direction at time k+1, T represents the time interval, and ω represents the angular velocity of the obstacle target.

[0048] (4) Current model; wherein the Current model is calculated according to the following formula:

[0049]

[0050] Among them, x k represents the displacement of the obstacle target in the x direction at time k, represents the speed of the obstacle target in the x direction at time k, represents the acceleration of the obstacle target in the x direction at time k, represents the jerk of the obstacle target in the x direction at time k, represents the mean of random acceleration, w(t) represents the mean of zero and the variance of maneuvering acceleration is σa 2 Gaussian white noise, α represents the inverse of the maneuver acceleration time constant. It should be noted that here It is a constant in each sampling interval. The maneuvering frequency α is usually determined based on experience and through actual measurement. It can be specifically set according to actual conditions, and the embodiment of the present invention does not limit this.

[0051] The discrete equation is as follows:

[0052]

[0053] Among them, x k represents the displacement of the obstacle target in the x direction at time k, represents the speed of the obstacle target in the x direction at time k, represents the acceleration of the obstacle target in the x direction at time k, x k+1 represents the displacement of the obstacle target in the x direction at time k+1, represents the speed of the obstacle target in the x direction at time k+1, represents the acceleration of the obstacle target in the x direction at time k+1, represents the mean of random acceleration, α represents the coefficient of Gaussian white noise and noise variance, T represents the time interval, and W(t) represents the discrete value of Gaussian white noise with zero mean.

[0054] (5) CM model; wherein the CM model is calculated according to the following formula:

[0055]

[0056] Among them, x k represents the displacement of the obstacle target in the x direction at time k, represents the speed of the obstacle target in the x direction at time k, y k represents the displacement of the obstacle target in the y direction at time k, represents the speed of the obstacle target in the y direction at time k, x k+1 represents the displacement of the obstacle target in the x direction at time k+1, represents the speed of the obstacle target in the x direction at time k+1, y k+1 represents the displacement of the obstacle target in the y direction at time k+1, represents the speed of the obstacle target in the y direction at time k+1, T represents the time interval, Δθ represents the change in the speed direction angle, and a t Indicates the tangential acceleration of curved motion, Δt k represents the time difference between time k+1 and time k, θ k Represents the velocity direction angle of the obstacle target at time k.

[0057] The change in the above velocity direction angle is calculated according to the following formula:

[0058] Δθ=θ k+1 -θ k (6)

[0059] Among them, Δθ represents the change in velocity direction angle, θ k+1 represents the velocity direction angle of the obstacle target at time k+1, θ k Represents the velocity direction angle of the obstacle target at time k.

[0060] Therefore, according to the above motion models pre-stored in the prior knowledge base, a target motion model set of obstacle targets can be constructed. One possible way to construct the target motion model set is as follows:

[0061] (1) Obtaining the collected information of the obstacle target; wherein the collected information includes but is not limited to type information, location information and parameter information; the specific settings can be made according to actual conditions, and the embodiments of the present invention do not limit this.

[0062] (2) Determine the first motion model set of the obstacle target based on the type information; specifically, determine the corresponding first motion model set according to the type information of the obstacle target. For example, large motor vehicles such as cars generally move at a uniform speed, at a uniform turning speed, and in a curve due to fixed gear positions. That is, when the type information of the obstacle target is a large motor vehicle, the corresponding first motion model set is {CV model, CT model, CM model}; for pedestrians, they generally move at a uniform speed of walking or sprinting in a straight line and in a current model. That is, when the type information of the obstacle target is a pedestrian, the corresponding first motion model set is {CV model, Current model}; for small motor vehicles such as bicycles or motorcycles, there will be certain acceleration and deceleration. That is, when the type information of the obstacle target is a small motor vehicle, the corresponding first motion model set is {CV model, CA model, Current model}, etc. Therefore, the first motion model set corresponding to each type of information can be specifically set according to the actual scenario.

[0063] (3) Determine a second motion model set of the obstacle target based on the position information; specifically, obtain the position information of the obstacle target based on the high-precision electronic map and the position of the obstacle target in the electronic map, such as a straight road, a U-turn lane, etc. Figure 3As shown. For motor vehicles, they must follow traffic rules. For example, on a straight lane, the corresponding second motion model set is {CV model, CA model, Current model}; on an overpass, corner lane or U-turn lane, the corresponding second motion model set is {CT model, Current model}; in a turning area or a variable road such as a fork road or ramp, the corresponding second motion model set is {CM model, Current model}, etc.; the specific settings can be made according to actual conditions.

[0064] (4) Determine a third motion model set of the obstacle target based on parameter information; wherein the parameter information includes at least one of the following: acquisition device information, obstacle target profile information, and meteorological information; wherein the acquisition device information is such as sensor information, etc., because the above parameter information is rarely considered in the existing methods, the third motion model set here defaults to the full set, that is, the third motion model set is {CV model, CA model, CT model, CM model, Current model}; the remaining non-full set situations can be set according to the actual scenario.

[0065] (5) Determine a target motion model set of the obstacle target based on the first motion model set, the second motion model set, and the third motion model set. Specifically, determine the intersection of the first motion model set, the second motion model set, and the third motion model set, and determine the intersection as the target motion model set of the obstacle target, so as to ensure that the target motion model set can accurately simulate the motion of the obstacle target.

[0066] For ease of understanding, a sensor is used as an example. Figure 4 As shown, the collected information includes sensor perception information, high-precision map information and other environmental information; wherein the sensor perception information includes sensor information and obstacle categories, and a sub-model set M1 of obstacle targets can be determined according to the sensor information, and a sub-model set M2 of obstacle targets can be determined according to the obstacle categories. Obstacle terrain can be matched and determined based on the sensor perception information and high-precision map information, and a sub-model set M3 of obstacle targets can be determined based on the obstacle terrain, and a sub-model set M4 of obstacle targets can be determined based on other perception information in other environmental information. n , where n is the number of collected information, and finally according to the above sub-model set M1, sub-model set M2, sub-model set M3 and sub-model set M n The intersection of determines the target motion model set of the obstacle target, that is, the target motion model set M = M1∩M2∩M3∩…M n .

[0067] It should be noted that the target motion model set of a specific obstacle target can be constructed according to the actual application scenario to ensure that the target motion model set can accurately simulate the motion of the obstacle target, and the embodiment of the present invention does not limit this.

[0068] In practical applications, since the target maneuver type, maneuver intensity, and maneuver time of the obstacle target are unknown, especially the obstacle target changes the motion mode many times during the maneuver, such as turning, accelerating, climbing, etc., the motion model is generally difficult to determine, that is, the target motion model of the obstacle target is difficult to match the actual motion mode of the obstacle target, which leads to filter divergence, reduced tracking accuracy, and even loss of the target. The embodiment of the present invention uses prior knowledge such as map environment information and type information of the obstacle target to determine the target motion model set of the obstacle target, thereby alleviating the problem that the existing obstacle target motion model is difficult to determine, and by constructing a priori knowledge base, the target motion model set of the obstacle target is also expanded, so that the target motion model set better fits the actual motion of the obstacle target, thereby improving the obstacle target state estimation accuracy.

[0069] Step S204, determining whether the target motion model set is consistent with the model set of the VSIMM filter at the previous moment;

[0070] Among them, the model set of the VSIMM filter at the previous moment includes the motion model corresponding to each filter in the VSIMM filter at the previous moment; specifically, it is determined whether each target motion model in the target motion model set is exactly the same as the motion model corresponding to each filter in the VSIMM filter at the previous moment; if so, it is determined that the target motion model set is consistent with the model set of the VSIMM filter at the previous moment, that is, no new filter model is activated in the model set of the VSIMM filter at the previous moment; if there is any difference, it is determined that the target motion model set is inconsistent with the model set of the VSIMM filter at the previous moment, that is, the model set of the VSIMM filter at the current moment is compared with the model set at the previous moment, and there is a new filter model. At this time, it is necessary to determine the model set of the VSIMM filter at the current moment, that is, adjust the filter model of the VSIMM filter.

[0071] Step S206, if not, determining the target model set of the VSIMM filter at the current moment based on the target motion model set and the model set of the VSIMM filter at the previous moment;

[0072] In practical applications, the adaptive filtering algorithm enables the target tracking system to make adaptive adjustments according to the maneuvering situation of the target, greatly improving the tracking accuracy. There are three main types of commonly used adaptive filtering algorithms: detection adaptive filtering algorithm, real-time identification adaptive filtering algorithm and multi-model method; among them, the detection adaptive filtering algorithm and the real-time identification adaptive filtering algorithm both correct the model or noise during the filtering process, but there will be general lag and the dependence on the model is relatively large, while the multi-model method such as the IMM (Interactive Multi-tude Model, interactive multi-model adaptive) filtering algorithm does not require maneuvering detection, but adjusts the probability of each model to achieve switching between different modes to achieve comprehensive adaptation. However, these algorithms do not essentially change the fixed structure of IMM, and still cannot solve the performance degradation problem caused by the interference of redundant target motion models in the IMM algorithm. The embodiments of the present invention utilize prior knowledge and external information of sensors to change the fixed structure of the IMM algorithm as much as possible, that is, to improve the filter performance through the VSIMM algorithm. For example, the motion model of the obstacle target can be dynamically increased or decreased, or the estimated position of the obstacle target can be corrected. Therefore, when the obstacle target maneuvers, the tracking filter in the VSIMM filter can converge more quickly, thereby obtaining higher tracking accuracy and improving the operation speed.

[0073] Specifically, since each target motion model in the target motion model set is not exactly the same as the motion model corresponding to each filter in the VSIMM filter at the previous moment, the filter model is adjusted in the target motion model set determined in the prior knowledge base according to certain rules. Specifically, first, the model set of the VSIMM filter at the current moment is obtained according to the union of the target motion model set and the model set of the VSIMM filter at the previous moment; then, the first model probability of the target motion model set and the second model probability of the model set of the VSIMM filter at the previous moment are obtained, and the ratio of the first model probability to the second model probability is obtained, and it is determined whether the ratio is greater than the first threshold. If so, the target model set of the VSIMM filter at the current moment is determined to be the target motion model set; if the ratio is not greater than the first threshold, it is determined whether the ratio is less than the second threshold. If so, it is determined that the target model set of the VSIMM filter at the current moment is the model set of the VSIMM filter at the previous moment. It should be noted that the specific values ​​of the first threshold and the second threshold can be set according to the actual situation, as long as the second threshold is less than the first threshold.

[0074] For ease of understanding, here we take the model set M of the VSIMM filter at the last moment, i.e., at moment k k and the target motion model set M m Take this as an example. Figure 5 As shown, according to Mk and M m Determine whether the model set of the VSIMM filter at the current time, i.e., time k+1, has been updated; if not, the model set of the VSIMM filter at time k+1 is M k+1 , and M k+1 =M k , if there is an update, then according to the target motion model set M m and the model set M of the VSIMM filter at time k k Determine the model set M of the VSIMM filter at time k+1 k+1 =M k ∪M m , and obtain the target motion model set M m The first model probability and the model set M of the VSIMM filter at time k k The second model probability Among them, the model probability of the model set is calculated according to the following formula:

[0075]

[0076] Among them, u M (k) represents the model probability of the model set M, m i represents the number of models in the model set M, u i (k) represents the probability of each model in the model ensemble.

[0077] Therefore, according to the above formula, the first model probability can be calculated and the second model probability The ratio of the first model probability to the second model probability is calculated according to the following formula:

[0078]

[0079] Where t represents the ratio of the probability of the first model to the probability of the second model, represents the target motion model set M at time k m The first model probability, The model set M represents the VSIMM filter at time k k The second model probability.

[0080] After the above ratio is calculated, it is determined whether the ratio t is greater than the first threshold t1. If so, the target model set of the VSIMM filter at time k+1 is determined to be the target motion model set, that is, M k+1 =M mIf not, determine whether the ratio t is less than the second threshold t2. If yes, determine that the target model set of the VSIMM filter at time k+1 is the model set of the VSIMM filter at time k, that is, M k+1 =M k , so according to the target motion model set M m and the model set u of the VSIMM filter at time k Mm (k), determine the target model set M of the VSIMM filter at time k+1 k+1 If the ratio t is greater than or equal to the second threshold t2 and less than or equal to the first threshold t1, continue to follow M k+1 =M k ∪M m Determine the target model set M of the VSIMM filter at time k+1 k+1 .

[0081] Step S208, obtaining the measurement information of the obstacle target at the current moment, so that the VSIMM filter estimates the filtering state according to the measurement information and the target model set, and obtains the total estimated output information of the VSIMM filter at the current moment;

[0082] For the VSIMM filter, the basic idea is: first, establish a total model set, which is composed of a series of multiple independent and compatible model sets. At each moment in the target tracking process, select a model set that best matches the target motion state from the total model set based on prior knowledge and the estimated state of the target; among them, each model set can be converted to each other according to certain rules, that is, each filter model in the model set is independent and can be converted to each other. Therefore, for the above-mentioned target model set, the VSIMM filter will also assign the corresponding target model to each filter in the VSIMM filter according to certain allocation rules, so that each filter performs filtering according to the assigned target model.

[0083] Specifically, the target filter in the corresponding VSIMM filter is determined according to each target model in the target model set; then, each target filter performs filter state estimation according to the measurement information and the corresponding target model to obtain the total estimated output information of the VSIMM filter at the current moment; the total estimated output information includes the total state estimation value and the total error covariance, the total state estimation value includes the state estimation value of each target filter at the current moment, and the total error covariance includes the error covariance of each target filter at the current moment. It should be noted that the measurement information of the obstacle target here includes but is not limited to the position information and speed information of the obstacle target, etc., which can be set according to the actual situation.

[0084] For ease of understanding, here we take the model set M of the VSIMM filter at the last moment, i.e., at moment kk and the target motion model set M m Take this as an example. If it is determined that the filter model of the VSIMM filter needs to be adjusted at time k+1, then for the target model set M of the VSIMM filter at time k+1, k+1 , an initial model probability can be assigned to the target model corresponding to each target filter and initialized, or the model probability of each target filter can be adjusted according to a transition probability matrix, wherein the transition probability matrix includes multiple transition probabilities, where the transition probability includes transition probabilities corresponding to multiple target filters, and the transition probability corresponding to each target filter also includes multiple transition probabilities between different target models corresponding to the target filter.

[0085] For the target model set M of the VSIMM filter at time k+1 k+1 , where the input of the target filter corresponding to the r target models at time k+1 includes: the mixed state estimate output at time k and the mixed error covariance. The mixed state estimate at time k can be calculated according to the following formula:

[0086]

[0087] in, represents the mixed state estimate at time k, represents the total state estimate of the target model i at time k, u i|j (k) represents the mixing probability of target model i to target model j at time k, where target model i is the model set M of the VSIMM filter at time k. k The target model j is the model set M of the VSIMM filter at time k+1 k+1 target model.

[0088] It should be noted that the above-mentioned mixed probability is the transfer probability from target model i to target model j at time k+1 determined based on the model probabilities of target model i and target model j, and the initial transfer probabilities of target model i and target model j. Therefore, by adjusting the transfer probability between models at each moment, the problem of uncertainty in the original transfer probability matrix due to the uncertainty of the obstacle target maneuver in the existing method is alleviated, thereby failing to well reflect the actual motion model switching of the obstacle target, thereby improving the tracking performance.

[0089] And, the mixed error covariance at time k is calculated according to the following formula:

[0090]

[0091] Among them, P 0j (k) represents the mixed error covariance at time k, P i(k) represents the total error covariance of target model i at time k, represents the total state estimate of the target model i at time k, represents the mixed state estimate at time k, u i|j (k) represents the mixing probability of target model i to target model j at time k, where target model i is the model set M of the VSIMM filter at time k. k The target model j is the model set M of the VSIMM filter at time k+1 k+1 target model.

[0092] Among them, the mixing probability can be calculated according to the following formula:

[0093]

[0094] Among them, u i|j (k) represents the mixing probability from target model i to target model j at time k, p ij represents the transition probability from target model i to target model j, u i (k) represents the model probability of target model i at time k, represents the predicted probability of target model j.

[0095] The predicted probability of target model j is calculated according to the following formula:

[0096]

[0097] in, represents the predicted probability of target model j, p ij represents the mixed probability of target model i transforming into target model j, u i (k) represents the model probability of target model i at time k, and target model i is the model set M of the VSIMM filter at time k. k The target model j is the model set M of the VSIMM filter at time k+1 k+1 target model.

[0098] At this time, according to the mixed state estimation and mixed error covariance at the above k moment, the target model set M k+1 Filter the target filter in the filter to obtain the total estimated output information of each target filter at time k+1, that is, the total state estimation value of each target filter at time k+1 is obtained and the total error covariance P j (k+1). And, the model probability of the target model j is updated based on the likelihood function of the target model j, wherein the likelihood function of the target model j is calculated according to the following formula:

[0099]

[0100] Among them, Λ j (k+1) represents the likelihood function of the jth target model, i.e., the target model j at time k+1, S j (k+1) represents the measurement covariance of target model j at time k+1, v j (k+1) represents the Kalman filter residual of the target model j at time k+1.

[0101] The model probability of the target model j at time k+1 is updated according to the following formula:

[0102]

[0103] Among them, u j (k+1) represents the model probability of target model j at time k+1, Λ j (k+1) represents the likelihood function of the target model j at time k+1, represents the predicted probability of the target model j, and C represents the model probability normalization constant.

[0104] Among them, the model probability normalization constant is calculated according to the following formula:

[0105]

[0106] Among them, C represents the model probability normalization constant, Λ j (k+1) represents the likelihood function of the target model j at time k+1, Represents the prediction probability of target model j, which is the model set M of the VSIMM filter at time k+1 k+1 target model.

[0107] According to the above formula, the target model set M of the VSIMM filter at time k+1 can be calculated k+1 The total estimated output information of . Among them, the total state estimate is calculated according to the following formula

[0108]

[0109] in, Represents the target model set M k+1 The total state estimate of represents the total state estimate of the target model j at time k+1, u j (k+1) represents the model probability of the target model j at time k+1.

[0110] And, the total error covariance is calculated according to the following formula:

[0111]

[0112] Among them, P(k+1) represents the target model set M k+1 The total error covariance, u j (k+1) represents the model probability of target model j at time k+1, P j (k+1) represents the total error covariance of target model j at time k+1, represents the total state estimate of the target model j at time k+1, Represents the target model set M k+1 The total state estimate.

[0113] For ease of understanding, an example is given here. Figure 6 As shown, based on the measurement information at time k+1, the target model set M of the VSIMM filter at time k+1 k+1 The first target filter output corresponding to the first target model has the first estimated output information and the first model probability, the second target filter output corresponding to the second target model has the second estimated output information and the second model probability, and so on, the Nth target filter output corresponding to the Nth target model has the Nth estimated output information and the Nth model probability, and the standard model set M can be obtained according to the first estimated output information, the second estimated output information to the Nth estimated output information. k+1 In addition, the model probability of the first model probability, the second model probability and the Nth model probability can be updated based on the likelihood function to improve the accuracy of the model probability corresponding to each target filter at each moment, thereby improving the target tracking performance.

[0114] Therefore, the transfer probability is adjusted by the mixed probability of the VSIMM filter, and the model probability of each target filter is updated by the likelihood function, thereby alleviating the problem that the original transfer probability matrix setting method in the VSIMM filter cannot well reflect the switching of the target's real motion model due to the uncertainty of the maneuver of the obstacle target; and due to the interference of a lot of external environmental information, the model probability of the obstacle target in different environments has also changed greatly, and the original setting method is difficult to obtain a more accurate model probability, resulting in a decrease in tracking performance. At the same time, it can also alleviate the problem that the measurement model of the sensor is difficult to determine and the measurement noise variance is constantly changing, which requires real-time estimation to ensure tracking accuracy, thereby improving the target tracking performance.

[0115] Step S210: Track the obstacle target according to the total estimated output information.

[0116] Therefore, the filtering method of the above-mentioned intelligent driving perception system makes full use of various prior knowledge of the intelligent driving perception system to constrain the motion model of the obstacle target, eliminates erroneous motion models, and thus determines the target motion model of the obstacle target, maximizes the utilization of prior knowledge information, and at the same time reduces the parallel calculation of a large number of motion models, and ensures that the target motion model of the obstacle target is more consistent with the actual movement of the obstacle target.

[0117] In addition, the VSIMM algorithm avoids the problem of tracking failure caused by divergence of traditional tracking filters, and alleviates the problem that the target model set of the filter in the general VSIMM algorithm is difficult to determine, resulting in reduced tracking accuracy and increased calculation time. By determining the target model set of the VSIMM filter at each moment, the calculation time of the VSIMM filter is reduced, and the state estimation accuracy of the VSIMM filter tracking filter is improved, thereby improving the tracking accuracy of obstacle targets, and then improving the performance of the intelligent driving perception system, which has good practical value.

[0118] Based on the above embodiment, the embodiment of the present invention further provides a filtering device for an intelligent driving perception system, which is applied to a server of the intelligent driving perception system, wherein the server is provided with a priori knowledge base and a variable structure interactive multiple model VSIMM filter. Figure 7 As shown, the device includes a first determination module 71, a judgment module 72, a second determination module 73, a filter estimation module 74 and a tracking module 75 connected in sequence; wherein the functions of each module are as follows:

[0119] A first determination module 71 is used to determine a target motion model set of the obstacle target based on the motion model in the prior knowledge base, wherein the motion model includes at least one of the following: a uniform velocity CV model, a uniform acceleration CA model, a uniform rotation speed CT model, a current model, and a curved motion CM model;

[0120] A judgment module 72 is used to judge whether the target motion model set is consistent with the model set of the VSIMM filter at the previous moment; wherein the model set of the VSIMM filter at the previous moment includes the motion model corresponding to each filter in the VSIMM filter at the previous moment;

[0121] A second determination module 73 is used to determine the target model set of the VSIMM filter at the current moment based on the target motion model set and the model set of the VSIMM filter at the previous moment if no;

[0122] The filter estimation module 74 is used to obtain the measurement information of the obstacle target at the current moment, so that the VSIMM filter estimates the filter state according to the measurement information and the target model set, and obtains the total estimated output information of the VSIMM filter at the current moment;

[0123] The tracking module 75 is used to track the obstacle target according to the total estimated output information.

[0124] The filtering device of the intelligent driving perception system provided by the embodiment of the present invention dynamically determines the target motion model of the obstacle target based on the motion model in the prior knowledge base, and determines the target model set of the VSIMM filter at the current moment according to the target motion model for filtering processing, so that the VSIMM filter can converge quickly when the obstacle target maneuvers, thereby improving the tracking accuracy of the obstacle target, thereby improving the performance of the intelligent driving perception system, and having good practical value.

[0125] In one possible embodiment, the judgment module 72 is also used to: judge whether each target motion model in the target motion model set is exactly the same as the motion model corresponding to each filter in the VSIMM filter at the previous moment; if so, judge that the target motion model set is consistent with the model set of the VSIMM filter at the previous moment; if there is any difference, judge that the target motion model set is inconsistent with the model set of the VSIMM filter at the previous moment.

[0126] In another possible embodiment, the second determination module 73 is also used to: obtain the model set of the VSIMM filter at the current moment based on the union of the target motion model set and the model set of the VSIMM filter at the previous moment; obtain the first model probability of the target motion model set and the second model probability of the model set of the VSIMM filter at the previous moment; and determine the target model set of the VSIMM filter at the current moment based on the ratio of the first model probability to the second model probability.

[0127] In another possible embodiment, the second determination module 73 is further used to: determine whether the ratio is greater than a first threshold value, and if so, determine that the target model set of the VSIMM filter at the current moment is the target motion model set; if not, determine whether the ratio is less than a second threshold value, and if so, determine that the target model set of the VSIMM filter at the current moment is the model set of the VSIMM filter at the previous moment; wherein the second threshold value is less than the first threshold value.

[0128] In another possible embodiment, the above-mentioned filter estimation module 74 is also used to: determine the target filter in the corresponding VSIMM filter according to each target model in the target model set; perform filtering state estimation on each target filter according to the measurement information and the corresponding target model to obtain the total estimated output information of the VSIMM filter at the current moment; wherein the total estimated output information includes a total state estimation value and a total error covariance, the total state estimation value includes the state estimation value of each target filter at the current moment, and the total error covariance includes the error covariance of each target filter at the current moment.

[0129] In another possible embodiment, each of the above target models is also configured with a model probability, and the device further includes: updating the model probability of each target model based on the likelihood function of each target model; wherein the likelihood function is calculated by the following formula:

[0130]

[0131] Among them, Λ j (k) represents the likelihood function of the jth target model at time k, S j (k) represents the measurement covariance of the jth target model at time k, v j (k) represents the Kalman filter residual of the j-th target model at time k.

[0132] In another possible embodiment, the first determination module 71 is further used to: obtain acquisition information of the obstacle target; wherein the acquisition information includes type information, location information and parameter information; determine a first motion model set of the obstacle target based on the type information; determine a second motion model set of the obstacle target based on the location information; determine a third motion model set of the obstacle target based on the parameter information; wherein the parameter information includes at least one of the following: acquisition device information, obstacle target profile information and meteorological information; determine a target motion model set of the obstacle target based on the first motion model set, the second motion model set and the third motion model set.

[0133] The filtering device of the intelligent driving perception system provided in the embodiment of the present invention has the same technical features as the filtering method of the intelligent driving perception system provided in the above embodiment, so it can also solve the same technical problems and achieve the same technical effects.

[0134] An embodiment of the present invention also provides an electronic device, including a processor and a memory, wherein the memory stores machine executable instructions that can be executed by the processor, and the processor executes the machine executable instructions to implement the filtering method of the above-mentioned intelligent driving perception system.

[0135] See also Figure 8As shown, the electronic device includes a processor 80 and a memory 81, wherein the memory 81 stores machine executable instructions that can be executed by the processor 80, and the processor 80 executes the machine executable instructions to implement the filtering method of the above-mentioned intelligent driving perception system.

[0136] Further, Figure 8 The electronic device shown further includes a bus 82 and a communication interface 83 , and the processor 80 , the communication interface 83 and the memory 81 are connected via the bus 82 .

[0137] Among them, the memory 81 may include a high-speed random access memory (RAM), and may also include a non-volatile memory (non-volatile memory), such as at least one disk storage. The communication connection between the system network element and at least one other network element is realized through at least one communication interface 83 (which can be wired or wireless), and the Internet, wide area network, local area network, metropolitan area network, etc. can be used. The bus 82 can be an ISA (Industrial Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus or an EISA (Enhanced Industry Standard Architecture) bus, etc. The above-mentioned bus can be divided into an address bus, a data bus, a control bus, etc. For the convenience of representation, Figure 8 Only one bidirectional arrow is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0138] The processor 80 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by the hardware integrated logic circuit or software instructions in the processor 80. The above processor 80 can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The methods, steps and logic block diagrams disclosed in the embodiments of the present invention can be implemented or executed. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor. The steps of the method disclosed in the embodiments of the present invention can be directly embodied as a hardware decoding processor to be executed, or the hardware and software modules in the decoding processor can be executed. The software module can be located in a mature storage medium in the field such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory 81, and the processor 80 reads the information in the memory 81 and completes the steps of the method of the above embodiment in combination with its hardware.

[0139] This embodiment also provides a machine-readable storage medium, which stores machine-executable instructions. When the machine-executable instructions are called and executed by a processor, the machine-executable instructions prompt the processor to implement the filtering method of the above-mentioned intelligent driving perception system.

[0140] The filtering method, device and computer program product of the intelligent driving perception system and electronic device provided in the embodiments of the present invention include a computer-readable storage medium storing program code. The instructions included in the program code can be used to execute the method described in the previous method embodiments. The specific implementation can be found in the method embodiments, which will not be repeated here.

[0141] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the system and device described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0142] In addition, in the description of the embodiments of the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0143] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a non-volatile computer-readable storage medium that is executable by a processor. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0144] In the description of the present invention, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc., indicating the orientation or positional relationship, are based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first", "second", and "third" are used for descriptive purposes only, and cannot be understood as indicating or implying relative importance.

[0145] Finally, it should be noted that the above-described embodiments are only specific implementations of the present invention, which are used to illustrate the technical solutions of the present invention, rather than to limit them. The protection scope of the present invention is not limited thereto. Although the present invention is described in detail with reference to the above-described embodiments, ordinary technicians in the field should understand that any technician familiar with the technical field can still modify the technical solutions recorded in the above-described embodiments within the technical scope disclosed by the present invention, or can easily think of changes, or make equivalent replacements for some of the technical features therein; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. A filtering method for an intelligent driving perception system, characterized in that: The method is applied to a server of the intelligent driving perception system, wherein the server is provided with a priori knowledge base and a variable structure interactive multiple model (VSIMM) filter, and the method comprises: Based on the motion models in the prior knowledge base, a target motion model set of the obstacle target is determined, wherein the motion model includes at least one of the following: a uniform velocity CV model, a uniform acceleration CA model, a uniform rotation speed CT model, a current model, and a curved motion CM model; Determine whether the target motion model set is consistent with the model set of the VSIMM filter at the previous moment; wherein the model set of the VSIMM filter at the previous moment includes the motion model corresponding to each filter in the VSIMM filter at the previous moment; If not, determining the target model set of the VSIMM filter at the current moment based on the target motion model set and the model set of the VSIMM filter at the previous moment; Acquire the measurement information of the obstacle target at the current moment, so that the VSIMM filter estimates the filtering state according to the measurement information and the target model set, and obtain the total estimated output information of the VSIMM filter at the current moment; Tracking the obstacle target according to the total estimated output information; The step of estimating the filtering state of the VSIMM filter according to the measurement information and the target model set includes: Determine a corresponding target filter in the VSIMM filter according to each target model in the target model set; By performing filtering state estimation on each target filter according to the measurement information and the corresponding target model, the total estimated output information of the VSIMM filter at the current moment is obtained; wherein the total estimated output information includes a total state estimation value and a total error covariance, the total state estimation value includes a state estimation value of each target filter at the current moment, and the total error covariance includes the error covariance of each target filter at the current moment.

2. The filtering method of the intelligent driving perception system according to claim 1, characterized in that: The step of determining whether the target motion model set is consistent with the model set of the VSIMM filter at the previous moment comprises: Determine whether each target motion model in the target motion model set is completely identical to the motion model corresponding to each filter in the VSIMM filter at the previous moment; If yes, it is determined that the target motion model set is consistent with the model set of the VSIMM filter at the previous moment; If there is any difference, it is determined that the target motion model set is inconsistent with the model set of the VSIMM filter at the previous moment.

3. The filtering method of the intelligent driving perception system according to claim 2, characterized in that: The step of determining the target model set of the VSIMM filter at the current moment based on the target motion model set and the model set of the VSIMM filter at the previous moment includes: Obtaining the model set of the VSIMM filter at the current moment according to the union of the target motion model set and the model set of the VSIMM filter at the previous moment; Obtaining a first model probability of the target motion model set and a second model probability of the model set of the VSIMM filter at a previous moment; A target model set of the VSIMM filter at the current moment is determined according to the ratio of the first model probability to the second model probability.

4. The filtering method of the intelligent driving perception system according to claim 3, characterized in that: The step of determining the target model set of the VSIMM filter at the current moment according to the ratio of the first model probability to the second model probability comprises: Determine whether the ratio is greater than a first threshold value, and if so, determine that the target model set of the VSIMM filter at the current moment is the target motion model set; If not, determine whether the ratio is less than a second threshold; if yes, determine that the target model set of the VSIMM filter at the current moment is the model set of the VSIMM filter at the previous moment; wherein the second threshold is less than the first threshold.

5. The filtering method of the intelligent driving perception system according to claim 1, characterized in that: Each of the target models is also configured with a model probability, and the method further comprises: Based on the likelihood function of each target model, the model probability of each target model is updated; wherein the likelihood function is calculated by the following formula: Among them, Λ j (k) represents the likelihood function of the jth target model at time k, S j (k) represents the measurement covariance of the jth target model at time k, v j (k) represents the Kalman filter residual of the j-th target model at time k.

6. The filtering method of the intelligent driving perception system according to claim 1, characterized in that: The step of determining a target motion model set of the obstacle target based on the motion model in the prior knowledge base includes: Acquire the acquisition information of the obstacle target; wherein the acquisition information includes type information, location information and parameter information; Determine a first motion model set of the obstacle target based on the type information; Determine a second motion model set of the obstacle target based on the position information; Determine a third motion model set of the obstacle target based on the parameter information; wherein the parameter information includes at least one of the following: acquisition device information, obstacle target profile information, and meteorological information; A target motion model set of the obstacle target is determined according to the first motion model set, the second motion model set and the third motion model set.

7. A filtering device for an intelligent driving perception system, characterized in that: The device is applied to a server of the intelligent driving perception system, wherein the server is provided with a priori knowledge base and a variable structure interactive multiple model (VSIMM) filter, and the device comprises: A first determination module is used to determine a target motion model set of the obstacle target based on the motion model in the prior knowledge base, wherein the motion model includes at least one of the following: a uniform velocity CV model, a uniform acceleration CA model, a uniform rotation speed CT model, a current model, and a curved motion CM model; A judgment module, used to judge whether the target motion model set is consistent with the model set of the VSIMM filter at the previous moment; wherein the model set of the VSIMM filter at the previous moment includes the motion model corresponding to each filter in the VSIMM filter at the previous moment; A second determination module, for determining, if no, a target model set of the VSIMM filter at a current moment based on the target motion model set and the model set of the VSIMM filter at a previous moment; A filter estimation module, used to obtain the measurement information of the obstacle target at the current moment, so that the VSIMM filter performs filter state estimation according to the measurement information and the target model set, and obtains the total estimated output information of the VSIMM filter at the current moment; A tracking module, configured to track the obstacle target according to the total estimated output information; The filtering estimation module is also used for: Determine a corresponding target filter in the VSIMM filter according to each target model in the target model set; By performing filtering state estimation on each target filter according to the measurement information and the corresponding target model, the total estimated output information of the VSIMM filter at the current moment is obtained; wherein the total estimated output information includes a total state estimation value and a total error covariance, the total state estimation value includes a state estimation value of each target filter at the current moment, and the total error covariance includes the error covariance of each target filter at the current moment.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the steps of the filtering method of the intelligent driving perception system described in any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the filtering method of the intelligent driving perception system described in any one of claims 1 to 6 are executed.

Images