Method, device and storage medium for determining motion state

Abstract

The present invention discloses a method, device and storage medium for determining a motion state, wherein the method for determining a motion state comprises: in the process of automatic driving of a vehicle, matching object information of a target object within a preset range from the vehicle to obtain attribute information corresponding to the object information of the target object; determining the state information of the target object according to at least one of a first preset method, a second preset method and a third preset method; wherein the first preset method comprises: inputting the attribute information of the target object into a data classification model, and determining the output result of the data classification model as the first state information, and the second preset method comprises: determining the second state information according to the attribute information of the target object and a preset convergence confidence threshold of the target object; the third preset method comprises: determining the third state information according to the attribute information and a preset motion logic of the target object; and determining whether the target object is in a stationary state based on the state information.

Description

Technical Field

[0001] The present invention relates to the technical field of hybrid electric vehicles, and in particular to a method, device and storage medium for determining a motion state. Background Art

[0002] With the rapid development of the global economy, the popularity of automobiles is getting higher and higher, and people's demand for the intelligence of autonomous driving technology is getting higher and higher. Similarly, there are also more stringent requirements for the safety of autonomous driving.

[0003] In autonomous driving, the driving system will inevitably have certain system errors, positioning errors, and detection errors of the perception algorithm. The noise caused by these factors interferes with the calculation results of the object speed, which will cause errors when the vehicle uses the perception algorithm to identify objects. For example, a stationary vehicle on the road will be identified as a moving vehicle. This error greatly affects the decision-making process of the back-end when planning the autonomous driving route. Figure 1 As shown in the figure, for a row of stationary vehicles on the side of the road, if the self-driving vehicle senses the presence of a moving vehicle A, then when planning and controlling, the self-driving vehicle will generate a predicted trajectory for vehicle A. After obtaining the predicted trajectory, the self-driving vehicle will adopt a strategy of sudden braking or rapid turning. These two strategies will cause the autonomous driving vehicle to run unsteadily, affecting driving safety and ride comfort, etc.

[0004] At present, most technical solutions to this problem are based on heuristic thinking, setting a simple filter judgment step for the speed of obstacles perceived by autonomous driving, that is, directly setting low-speed obstacles with a speed less than a certain threshold to a stationary state. However, this method not only requires a large amount of data for debugging, is time-consuming and labor-intensive, and is not robust enough, but also has the risk of misjudging the speed of truly low-speed moving objects as stationary. The driving route planned under this misjudgment seriously affects the safety of autonomous driving vehicle operation.

[0005] Therefore, no effective solution has been proposed so far for the problem of inaccurate object state determination during autonomous driving in the prior art. Summary of the invention

[0006] In view of this, an object of the present invention is to provide a method, device and storage medium for determining a motion state, so as to at least solve the technical problem of inaccurate object state determination during autonomous driving in the prior art.

[0007] In order to achieve the above-mentioned purpose, in a first aspect, the present invention provides a method for determining a motion state, the method comprising: in the process of automatic driving of a vehicle, matching object information of a target object within a preset range from the vehicle to obtain attribute information corresponding to the object information of the target object; determining the state information of the target object according to at least one of a first preset method, a second preset method, and a third preset method, the state information comprising at least one of the following: first state information, second state information, and third state information; wherein, the first preset method comprises: inputting the attribute information of the target object into a data classification model, and determining the output result of the data classification model as the first state information, the second preset method comprises: determining the second state information according to the attribute information of the target object and a preset convergence confidence threshold of the target object; the third preset method comprises: determining the third state information according to the attribute information and a preset motion logic of the target object; and determining whether the target object is in a stationary state based on the state information.

[0008] In some embodiments, object information of a target object within a preset range from the vehicle is matched to obtain attribute information corresponding to the object information of the target object, including: obtaining a first position value of the target object in the object information and a first size value of the target object in the object information; searching for first feature information matching the object information from a pre-stored feature information table, wherein the first feature information includes a second position value of the target object and a second size value of the target object, and a difference between the first position value and the second position value is less than a first threshold value and / or a difference between the first size value and the second size value is less than a second threshold value; obtaining second feature information corresponding to the first feature information from the feature information table, the second feature information including at least one of the following: speed information of the target object, direction information of the target object, variance information of the target object, association confidence information of the target object, and detection confidence information of the target object; and determining the second feature information as the attribute information corresponding to the object information of the target object.

[0009] In some embodiments, the attribute information of the target object is input into a data classification model, and the output result of the data classification model is determined as the first state information of the target object, including: the speed information of the target object, the direction information of the target object, the variance information of the target object, the association confidence information of the target object, and the detection confidence information of the target object in the attribute information of the target object are simultaneously input into the data classification model, and the data classification model is used to determine the state information of the target object; the output result of the data classification model is determined as the first state information of the target object, wherein the first state information includes one of the following: action state, static state.

[0010] In some embodiments, the second state information of the target object is determined based on the attribute information of the target object and a preset convergence confidence threshold of the target object, including: parsing the detection speed and the filtered speed of the target object from the speed information of the target object, the filtered speed representing the speed after noise filtering of the detection speed of the target object; calculating the speed noise value of the target object based on the detection speed of the target object and the filtered speed of the target object; determining the convergence confidence corresponding to the speed information of the target object based on the speed noise value; and determining the convergence confidence corresponding to the speed information of the target object based on the convergence confidence and The preset convergence confidence threshold determines the second state information of the target object, wherein the second state information of the target object includes one of the following: a speed value convergence state, a speed value non-convergence state; determining the second state information of the target object according to the convergence confidence and the preset convergence confidence threshold includes: when the convergence confidence is greater than the preset convergence confidence threshold, determining that the second state information of the target object is a speed value convergence state; when the convergence confidence is not greater than the preset convergence confidence threshold, determining that the second state information of the target object is a speed value non-convergence state.

[0011] In some embodiments, determining a convergence confidence level corresponding to the speed information of the target object according to the speed noise value includes: calculating a first speed convergence value corresponding to the speed information of the target object at a first moment according to a preset convergence noise threshold and the speed noise value, wherein the first moment represents an initial moment; and determining the first speed convergence value as a first convergence confidence level of the speed information of the target object at the first moment.

[0012] In some embodiments, determining a convergence confidence corresponding to the speed information of the target object according to the speed noise value includes: calculating a historical speed noise value of the target object according to a historical detection speed of the target object and a filtered speed of the target object; calculating a historical speed convergence value corresponding to the historical speed information of the target object based on a preset convergence noise threshold and the historical speed noise value; setting the second state information of the target object at a first moment as the second state information of the target object at a second moment, wherein the second moment is after the first moment; when the second state information of the target object at the second moment is a speed value convergence state, determining an average value among the historical speed convergence values ​​as a second convergence confidence of the speed information of the target object at the second moment; and when the second state information of the target object at the second moment is a speed value non-convergence state, determining a minimum value among the historical speed convergence values ​​as the second convergence confidence of the speed information of the target object at the second moment.

[0013] In some embodiments, determining the third state information of the target object according to the attribute information of the target object and the preset motion logic of the target object includes: determining the motion direction of the target object within a preset time period according to the attribute information of the target object; in the case where it is determined that the motion direction is unidirectional, determining multiple groups of motion time periods from the preset time period, each group of motion time periods in the multiple groups of motion time periods includes a first time period and a second time period, the end time of the first time period in each group is the start time of the second time period, and the duration of the first time period is the same as the duration of the second time period; obtaining a first moving distance of the target object in the first time period of each group of motion time periods, and obtaining a first moving distance of the target object in the first time period of each group of motion time periods. a second moving distance within a second time period, and determining a first number of movement time periods in which the first moving distance is greater than the second moving distance, and a second number of movement time periods in which the first moving distance is not greater than the second moving distance; determining third state information of the target object according to the first number and the second number, wherein the third state information of the target object includes one of the following: action state, static state; determining the third state information of the target object according to the first number and the second number includes: when the first number is less than the second number, determining that the third state information of the target object is a static state; when the first number is not less than the second number, determining that the third state information of the target object is a moving state.

[0014] In some embodiments, determining whether the target object is in a stationary state based on the state information includes: determining the motion state of the target object based on the first state information, the second state information and the third state information, including: in the case where the first state information is determined to be a stationary state, the second state information is a velocity value convergence state and the third state information is a stationary state, determining that the state information of the target object is a stationary state; in the case where any one of determining that the first state information is a stationary state, the second state information is a velocity value convergence state and the third state information is a stationary state is not established, determining that the state information of the target object is a motion state.

[0015] In a second aspect, the present invention also provides a device for determining a motion state, the device comprising: an information matching module, which is used to match the object information of a target object within a preset range from the vehicle during automatic driving of the vehicle, and obtain attribute information corresponding to the object information of the target object; an information determination module, which is used to determine the state information of the target object according to at least one of a first preset method, a second preset method, and a third preset method, and the state information includes at least one of the following: first state information, second state information, and third state information; wherein, the first preset method comprises: inputting the attribute information of the target object into a data classification model, and determining the output result of the data classification model as the first state information, and the second preset method comprises: determining the second state information according to the attribute information of the target object and a preset convergence confidence threshold of the target object; the third preset method comprises: determining the third state information according to the attribute information and a preset motion logic of the target object; a state determination module, which is used to determine whether the target object is in a stationary state based on the state information.

[0016] In a third aspect, the present invention also provides a computer-readable storage medium, which includes a stored program, wherein when the program is running, the device where the computer-readable storage medium is located is controlled to execute the method for determining the motion state described in any of the above technical solutions.

[0017] The present invention obtains attribute information corresponding to the object information of the target object by matching the object information of the target object within a preset range of the vehicle during the process of automatic driving of the vehicle; determines the state information of the target object according to at least one of a first preset method, a second preset method, and a third preset method, and the state information includes at least one of the following: first state information, second state information, and third state information; wherein, the first preset method includes: inputting the attribute information of the target object into a data classification model, and determining the output result of the data classification model as the first state information, and the second preset method includes: determining the second state information according to the attribute information of the target object and a preset convergence confidence threshold of the target object; the third preset method includes: determining the third state information according to the attribute information and the preset motion logic of the target object; determining whether the target object is in a stationary state based on the state information can obtain a more accurate object state, making automatic driving more efficient and reliable.

[0018] 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

[0019] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0020] Figure 1 is a schematic diagram of vehicle trajectory prediction in the related art;

[0021] Figure 2 is a flow chart of a method for determining a motion state provided by the present invention;

[0022] Figure 3 It is a schematic diagram of the framework of the object state judgment logic provided by the present invention;

[0023] Figure 4 is a schematic diagram of the preset motion logic provided by the present invention (I);

[0024] Figure 5 is a schematic diagram of the preset motion logic provided by the present invention (II);

[0025] Figure 6 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0026] The specific embodiments of the present invention are described in detail below with reference to the accompanying drawings, but are not intended to limit the present invention.

[0027] It should be understood that various modifications may be made to the embodiments disclosed herein. Therefore, the above description should not be considered as limiting, but only as an example of an embodiment. Other modifications within the scope and spirit of the invention will occur to those skilled in the art.

[0028] The accompanying drawings, which are incorporated in and constitute a part of the specification, illustrate embodiments of the invention and, together with the general description of the invention given above and the detailed description of the embodiments given below, serve to explain the principles of the invention.

[0029] These and other characteristics of the invention will become apparent from the following description of a preferred form of embodiment given as a non-limiting example, with reference to the accompanying drawings.

[0030] It should also be understood that, although the invention has been described with reference to some specific examples, those skilled in the art will be able to realize many other equivalent forms of the invention that have the characteristics recited in the claims and are therefore within the scope of protection defined thereby.

[0031] The above and other aspects, features and advantages of the present invention will become more apparent from the following detailed description when taken in conjunction with the accompanying drawings.

[0032] Specific embodiments of the present invention are described hereinafter with reference to the accompanying drawings; however, it should be understood that the disclosed embodiments are merely examples of the present invention, which may be implemented in a variety of ways. Well-known and / or repeated functions and structures are not described in detail to avoid unnecessary or redundant details that obscure the present invention. Therefore, the specific structural and functional details disclosed herein are not intended to be limiting, but merely serve as a basis and representative basis for the claims to teach those skilled in the art to use the present invention in a variety of ways with substantially any suitable detailed structure.

[0033] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0034] This specification may use the phrases "in one embodiment," "in another embodiment," "in a further embodiment," or "in other embodiments," which may all refer to one or more of the same or different embodiments according to the present invention.

[0035] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments.

[0036] The first aspect of the present invention provides a method for determining a motion state. The method is suitable for matching object information of a target object within a preset range from the vehicle during automatic driving of a vehicle, obtaining attribute information corresponding to the object information of the target object, and determining the state information of the target object according to a plurality of preset methods to obtain a more accurate object state, thereby making automatic driving more efficient and reliable.

[0037] like Figure 2 As shown, according to the method for determining the motion state of the present invention, the steps are described in detail as follows:

[0038] Step S202: During the process of automatic driving of the vehicle, object information of a target object within a preset range from the vehicle is matched to obtain attribute information corresponding to the object information of the target object.

[0039] The target object within the preset range of the vehicle may include other vehicles within the preset range of the vehicle, and may also include obstacles within the preset range of the vehicle.

[0040] Step S204: Determine the state information of the target object at least according to one of the first preset method, the second preset method, and the third preset method, the state information including at least one of the following: first state information, second state information, and third state information; wherein, the first preset method includes: inputting the attribute information of the target object into a data classification model, and determining the output result of the data classification model as the first state information, the second preset method includes: determining the second state information according to the attribute information of the target object and a preset convergence confidence threshold of the target object; the third preset method includes: determining the third state information according to the attribute information and a preset motion logic of the target object.

[0041] Step S206: Determine whether the target object is in a stationary state based on the state information.

[0042] Through the above steps, during the process of automatic driving of the vehicle, the object information of the target object within the preset range of the vehicle is matched to obtain the attribute information corresponding to the object information of the target object; the state information of the target object is determined according to at least one of the first preset method, the second preset method, and the third preset method; wherein the first preset method includes: inputting the attribute information of the target object into a data classification model, and determining the output result of the data classification model as the first state information, the second preset method includes: determining the second state information according to the attribute information of the target object and the preset convergence confidence threshold of the target object; the third preset method includes: determining the third state information according to the attribute information and the preset motion logic of the target object; determining whether the target object is in a stationary state based on the state information can obtain a more accurate object state, making automatic driving more efficient and reliable.

[0043] In step S202, object information of a target object within a preset range from the vehicle is matched to obtain attribute information corresponding to the object information of the target object, specifically including:

[0044] Acquire a first position value of the target object in the object information and a first size value of the target object in the object information; search for first feature information matching the object information from a pre-stored feature information table, wherein the first feature information includes a second position value of the target object and a second size value of the target object, and a difference between the first position value and the second position value is less than a first threshold value and / or a difference between the first size value and the second size value is less than a second threshold value; acquire second feature information corresponding to the first feature information from the feature information table, the second feature information including at least one of the following: speed information of the target object, direction information of the target object, variance information of the target object, association confidence information of the target object, and detection confidence information of the target object; determine the second feature information as attribute information corresponding to the object information of the target object.

[0045] Before executing the above steps, the image of the obstacle (i.e., the target object) can be annotated to obtain true value annotation data (which can be referred to as annotation data for short), where the true value annotation data may include the position information of the target object, the size information of the target object, and the dynamic and static attributes of the target object. Then, the collected obstacle image is processed according to the visual perception algorithm commonly used in the field of autonomous driving to obtain the shallow target feature information of the obstacle, and obstacle matching is performed on the shallow target feature information of the obstacle and the true value annotation data.

[0046] In one embodiment, the above scheme can be further described in conjunction with Tables 1 and 2. Table 1 represents the true value annotation data of the obstacle, that is, the object information of the target object. Table 2 is a shallow target feature information table of the obstacle, which is equivalent to the above-mentioned pre-stored feature information table. Among them, Tables 1 and 2 can provide the data of the training set required by the random forest tree model. In this embodiment, the random forest tree model can be trained based on the state in the dynamic and static separation influencing factor symbol 1 in Table 1 and the shallow target feature information extracted in Table 2. After the training is completed, a part of the data set is extracted for testing. The shallow target feature information extracted in Table 2 includes Vx, Vy, Ax, Ay, which is equivalent to the speed information of the above-mentioned target object, and Dx, Dy, which is equivalent to the direction information of the above-mentioned target object, σx, σy, σvx, σvy, σax, σay, σdx, σdy, the variance information of the target object, and conf, which is equivalent to the association confidence information of the above-mentioned target object, and score, which is equivalent to the detection confidence information of the above-mentioned target object) to perform a data classification model (such as a random forest tree model).

[0047] Then, find the dynamic and static separation influencing factor symbol 2 (equivalent to the above-mentioned first feature information, including x2, y2, z2, l2, w2, h2) that matches the dynamic and static separation influencing factor symbol 1 in Table 1 from Table 2, and then use the state in Table 1 and Vx, Vy, Ax, Ay, Dx, Dy, σx, σy, σvx, σvy, σax, σay, σdx, σdy, conf, and score in Table 2 as training samples of the random forest tree model to train the random forest tree model. After successful training, the corresponding state can be obtained by inputting Vx, Vy, Ax, Ay, Dx, Dy, σx, σy, σvx, σvy, σax, σay, σdx, σdy, conf, and score into the random forest tree model.

[0048] Table 1: Annotated data table (only contains the size, position, and dynamic and static properties of the true value)

[0049] Symbol of factors affecting dynamic and static separation 1 Meaning of symbols <![CDATA[x1]]> <![CDATA[X1 coordinate of the center point of the obstacle]]> <![CDATA[y1]]> <![CDATA[Y1 coordinate of the center point of the obstacle]]> <![CDATA[z1]]> <![CDATA[Z1 coordinate of the center point of the obstacle]]> <![CDATA[l1]]> Obstacle length <![CDATA[w1]]> Obstacle width <![CDATA[h1]]> Obstacle height state Obstacle dynamic and static properties

[0050] Table 2: Shallow target feature information

[0051]

[0052]

[0053] For finding the first feature information matching the object information from the pre-stored feature information table, the IOU (Intersection over Union, a standard for measuring the accuracy of detecting corresponding objects in a specific data set, any task that obtains a prediction range (bounding boxes) in the output can be measured by IoU) data association method can be used to achieve it, and the annotated data x1, y1, z1, l1, w1, h1 of the same frame and the shallow target feature information x2, y2, z2, l2, w2, h2 stored online are matched. When the IOU is greater than a certain threshold, the two objects are considered to match.

[0054] The x1, y1, and z1 in Table 1 are equivalent to the first position value, the l1, w1, and h1 in Table 1 are equivalent to the first size value, the x2, y2, and z2 in Table 2 are equivalent to the second position value, and the l2, w2, and h2 in Table 2 are equivalent to the second size value. Taking x1 as the first position value as an example, if the difference between x1 and x2 is less than the preset difference of the center point x coordinate (equivalent to the first threshold), it can be determined that x1 and x2 are matched successfully. Among them, one or more of x1, y1, and z1 can also be taken as the first position value as an example, and the present invention will not be repeated here.

[0055] Similarly, the matching process between the first size value and the second size value may refer to the matching process between the first position value and the second position value, which will not be elaborated in detail in the present invention.

[0056] In one embodiment, the data pairs matching Table 1 and Table 2 can be calculated in the following manner. Assuming that the kth frame has i labeled data in Table 1 and j online stored data in Table 2 (i.e. shallow target feature information), the following matching matrix Mk can be finally obtained:

[0057]

[0058] Among them, m ij Represents the IOU value between the i-th labeled data and the j-th online stored data. Continuing to use the GNN matching algorithm on the above matrix, we can get the matched data pairs in Table 1 and Table 2. At this time, we can get the data corresponding to the dynamic and static separation influencing factor symbol 2 in Table 2, which has a different meaning from the dynamic and static separation influencing factor symbol 1 in Table 1, namely Vx, Vy, Ax, Ay, Dx, Dy, σx, σy, σvx, σvy, σax, σay, σdx, σdy, conf, score. These data represent the attribute information corresponding to the object information of the target object.

[0059] Before training the model, first, the attribute information of the target object is screened, for example, the speed information of the target object is screened according to a preset speed range of the target object, or the displacement information of the target object is screened according to a preset displacement range of the target object, so as to obtain data that meets the preset speed range and the preset displacement range.

[0060] Next, the data is normalized. For example, the normalization formulas for Vx and Vy are:

[0061]

[0062] The same goes for the remaining variables.

[0063] The principle of determining the state information of the target object based on the random forest tree model includes: classifying the data to obtain multiple sample sets, that is, obtaining a sample set for each type of data in Vx, Vy, Ax, Ay, Dx, Dy, σx, σy, σvx, σvy, σax, σay, σdx, σdy, conf, and score, and establishing a classifier for each sample set to obtain m classifiers, where m is a positive integer. The above data is input into the m classifiers, and the classification result of the data is obtained according to the voting results of the m classifiers, that is, according to the data of Vx, Vy, Ax, Ay, Dx, Dy, σx, σy, σvx, σvy, σax, σay, σdx, σdy, conf, and score of the target object, it is determined whether the target object is in a stationary state.

[0064] In step S204, the process of determining the state information of the target object based on the first preset method is to input the attribute information of the target object into a data classification model, and determine the output result of the data classification model as the first state information of the target object.

[0065] Specifically, it may include: simultaneously inputting the speed information of the target object, the direction information of the target object, the variance information of the target object, the association confidence information of the target object, and the detection confidence information of the target object in the attribute information of the target object into the data classification model, and the data classification model is used to determine the state information of the target object; and determining the output result of the data classification model as the first state information of the target object, wherein the first state information includes one of the following: action state, static state.

[0066] In step S204, in the scheme of obtaining the state information of the target object using the second preset method, the object information of the target object within the preset range from the vehicle is matched to obtain the attribute information corresponding to the object information of the target object.

[0067] First, step S1: parse the detection speed and filtered speed of the target object from the speed information of the target object, the filtered speed represents the speed after noise filtering of the detection speed of the target object; calculate the speed noise value of the target object according to the detection speed of the target object and the filtered speed of the target object; determine the convergence confidence corresponding to the speed information of the target object according to the speed noise value.

[0068] Secondly, step S2: determining the second state information of the target object according to the convergence confidence and the preset convergence confidence threshold, wherein the second state information of the target object includes one of the following: a speed value convergence state, a speed value non-convergence state; determining the second state information of the target object according to the convergence confidence and the preset convergence confidence threshold includes: when the convergence confidence is greater than the preset convergence confidence threshold, determining that the second state information of the target object is a speed value convergence state; when the convergence confidence is not greater than the preset convergence confidence threshold, determining that the second state information of the target object is a speed value non-convergence state.

[0069] Specifically: in step S1, there are the following schemes for determining the convergence confidence corresponding to the speed information of the target object according to the speed noise value:

[0070] First, step a. Calculate the velocity noise value Error V :Error V =Meas V -Filter V .

[0071] Among them, Meas V It indicates the detection speed of the target object. V It represents the speed after noise filtering of the detection speed of the target object.

[0072] Step b. Set the convergence score threshold and initialize the convergence noise threshold Thres V .

[0073] Then, step c is executed to calculate the convergence confidence of the target object, which specifically includes:

[0074] First, calculate the first convergence confidence of the first frame velocity of the target object:

[0075] Specifically: a first speed convergence value corresponding to the speed information of the target object at a first moment is calculated according to a preset convergence noise threshold and the speed noise value, wherein the first moment represents an initial moment; and the first speed convergence value is determined as a first convergence confidence of the speed information of the target object at the first moment.

[0076] Among them, for the first speed convergence value, the convergence score (equivalent to the first convergence confidence mentioned above) is calculated according to the following formula: V ,

[0077]

[0078] Second, calculate the second convergence confidence of the multi-frame velocity of the target object:

[0079] Specifically, a historical speed noise value of the target object is calculated according to a historical detection speed of the target object and a filtered speed of the target object; a historical speed convergence value corresponding to the historical speed information of the target object is calculated based on a preset convergence noise threshold and the historical speed noise value; the second state information of the target object at a first moment is set as the second state information of the target object at a second moment, wherein the second moment is after the first moment; when the second state information of the target object at the second moment is a speed value convergence state, an average value among the historical speed convergence values ​​is determined as a second convergence confidence of the speed information of the target object at the second moment; when the second state information of the target object at the second moment is a speed value non-convergence state, a minimum value among the historical speed convergence values ​​is determined as the second convergence confidence of the speed information of the target object at the second moment.

[0080] For example, if the target object's speed data is multiple frames, the following formula can be used to calculate the convergence score of each frame other than the first frame (equivalent to the second convergence confidence mentioned above): V .

[0081] First, calculate the historical speed noise value Errorhistory V ,

[0082] Errorhistory V =Meashistory V -Filter V .

[0083] Among them, Meashistory V It indicates the historical detection speed of the target.

[0084] At the same time, the convergence score table is calculated, which indicates the convergence score of each frame other than the first frame:

[0085]

[0086] Set the current target speed convergence to be equal to the previous target speed convergence, and calculate the convergence confidence Convergenceconf V .

[0087] The current goal does not converge, find the minimum value of the convergence score table:

[0088] Convergenceconf V =min(Convergencescore V ).

[0089] The current target converges, and the average value of the convergence score table is solved:

[0090] Convergenceconf V =mean(Convergencescore V ).

[0091] Furthermore, in step S2, it also includes:

[0092] Step d. Update the current target speed convergence.

[0093] 1). Determine the preset convergence confidence threshold Convergenceconfthres V .

[0094] 2). If the calculated convergence confidence is greater than the convergence confidence threshold, the current target speed converges, i.e. Convergenceconf V >Convergenceconfthres V .

[0095] 3). If the calculated convergence confidence is less than the convergence confidence threshold, the current target speed does not converge, that is, Convergenceconf V <Convergenceconfthres V .

[0096] Through the above calculation, the convergence of the current target speed can be obtained.

[0097] The present invention makes a convergence judgment based on the change of speed in the above steps, judges whether the speed difference between the previous and next frames is too large, and then compares the historical multiple frames with the threshold change to finally obtain the convergence of the target speed.

[0098] In step S204, in the scheme of obtaining the state information of the target object using the third preset method, the third state information of the target object can be determined according to the attribute information of the target object and the preset motion logic of the target object, specifically:

[0099] First, the movement direction of the target object within a preset time period is determined according to the attribute information of the target object; when it is determined that the movement direction is unidirectional, a plurality of groups of movement time periods are determined from the preset time period, each group of movement time periods in the plurality of movement time periods includes a first time period and a second time period, the end time of the first time period in each group is the start time of the second time period, and the duration of the first time period is the same as the duration of the second time period; a first moving distance of the target object in the first time period of each group of movement time periods and a second moving distance of the target object in the second time period of each group of movement time periods are obtained, and a first number of movement time periods in which the first moving distance is greater than the second moving distance and a second number of movement time periods in which the first moving distance is not greater than the second moving distance are determined.

[0100] The movement direction of the target object within the preset time period includes unidirectional and multi-directional, and multi-directional can be understood as at least two directions.

[0101] Secondly, the third state information of the target object is determined according to the first number and the second number, wherein the third state information of the target object includes one of the following: action state, static state. The third state information of the target object is determined according to the first number and the second number, including: when the first number is not less than the second number, determining that the third state information of the target object is a motion state; when the first number is less than the second number, determining that the third state information of the target object is a static state.

[0102] like Figure 4As shown, the displacement distance of the target object from the (t-2) frame to the (t-1) frame includes: ΔDis(t-2,t-1), ΔDis(t-2,t), ΔDis(t-2,t), ΔDis(t-2,t+1), ΔDis(t-2,t+1), ΔDis(t-2,t+2). In the same direction of motion, ΔDis(t-2,t-1)<ΔDis(t-2,t), ΔDis(t-2,t)<ΔDis(t-2,t+1), ΔDis(t-2,t+1)<ΔDis(t-2,t+2). It can be determined that the target object is moving in one direction continuously in several consecutive frames, which is a moving state. Among them, ΔDis(t-2,t-1) represents the displacement difference from the t-2 frame to the t-1 frame, and the same applies to the other distances.

[0103] like Figure 5 As shown, in the same direction of motion, ΔDis(t-2,t-1)>ΔDis(t-2,t), ΔDis(t-2,t)<ΔDis(t-2,t+1), ΔDis(t-2,t+1)>ΔDis(t-2,t+2). At this time, the number of times the distance is large is less than the number of times the distance is small, and the object is determined to be stationary.

[0104] In addition, when it is determined that the movement direction is multi-directional, the third state information of the target object can be directly determined to be a stationary state.

[0105] The present invention determines whether an obstacle is stationary based on a preset motion logic, further enriches the types of solutions for determining whether an obstacle is stationary, and improves the accuracy of the determination result.

[0106] In step S206, it is determined whether the target object is in a stationary state based on the state information. There are multiple implementation schemes:

[0107] First, the motion state of the target object is determined based on the first state information, the second state information and the third state information, including: when the first state information is determined to be a stationary state, the second state information is a velocity value convergence state and the third state information is a stationary state, the state information of the target object is determined to be a stationary state; when any one of the first state information being a stationary state, the second state information being a velocity value convergence state and the third state information being a stationary state is not established, the state information of the target object is determined to be a motion state.

[0108] like Figure 5As shown, the surrounding environment can be perceived based on the radar automatic driving perception and tracking algorithm, and the perception result can be output. A shallow target feature information table can be extracted from the perception result. The input data of the random forest tree model can be matched from the shallow target feature information table according to the labeled data to obtain the output result of the random forest tree model, and the convergence of the speed can be judged, and the motion state can be judged according to the preset motion logic. If the output result of the random forest tree is a stationary state, the convergence of the speed is the convergence of the speed value, and the result of the preset motion logic judgment is a stationary state, then the target object is judged to be in a stationary state. Otherwise, the target object is judged to be in a moving state.

[0109] The present invention adopts the output results of the autonomous driving perception tracking algorithm and the true value of the labeled data, and uses logical judgments based on random forest trees and convergence to solve the problem of inaccurate existing perception solutions. First, the shallow target feature information of the obstacle is extracted, and the training data of the random forest tree model is obtained according to the labeled true value data of the obstacle, and the random forest tree model is trained. The dynamic and static state of the obstacle is obtained according to the trained random forest tree model, and then the convergence of the target speed is obtained according to the change of speed, and the dynamic and static state of the obstacle is judged by the preset motion logic. Finally, combined with the judgment results of the above process, it is determined whether the obstacle is in a static state. This step-by-step judgment process fully utilizes the advantages of the strong generalization ability of the random forest tree model. For unbalanced data sets, it can also play a role in balancing errors, and construct a set of rigorous judgment principles for obstacle state information, which not only improves the accuracy of the obstacle state information perception, but also realizes the dynamic and static separation of obstacles, which can further improve the safety during the autonomous driving process and improve the user's driving experience.

[0110] Second, determining the motion state of the target object based on the first state information and the second state information, including: when the first state information is determined to be a stationary state and the second state information is determined to be a velocity value convergence state, determining the state information of the target object to be a stationary state; when either the first state information is determined to be a stationary state or the second state information is determined to be a velocity value convergence state is not established, determining the state information of the target object to be a motion state.

[0111] In addition to the above scheme, in some embodiments, other schemes are also included, specifically: determining the motion state of the target object based on the first state information and the third state information, determining the motion state of the target object based on the second state information and the third state information, determining the motion state of the target object based on the first state information, determining the motion state of the target object based on the second state information, and determining the motion state of the target object based on the third state information. These schemes can be implemented with reference to the above scheme of determining the motion state of the target object based on the first state information, the second state information, and the third state information, and the present invention will not repeat them here.

[0112] After completing the above steps, during the process of automatic driving of the vehicle, the object information of the target object within a preset range from the vehicle is matched to obtain attribute information corresponding to the object information of the target object; and the state information of the target object is determined according to at least one of the first preset method, the second preset method, and the third preset method; based on the state information obtained in multiple preset methods, it is determined whether the target object is in a stationary state, so as to obtain a more accurate object state, thereby making automatic driving more efficient and reliable.

[0113] Example 2

[0114] The second aspect of the present invention also provides a device for determining a motion state, the device comprising: an information matching module, which is used to match the object information of a target object within a preset range from the vehicle during automatic driving of the vehicle, and obtain attribute information corresponding to the object information of the target object; an information determination module, which is used to determine the state information of the target object according to at least one of a first preset method, a second preset method, and a third preset method, and the state information includes at least one of the following: first state information, second state information, and third state information; wherein the first preset method comprises: inputting the attribute information of the target object into a data classification model, and determining the output result of the data classification model as the first state information, and the second preset method comprises: determining the second state information according to the attribute information of the target object and a preset convergence confidence threshold of the target object; the third preset method comprises: determining the third state information according to the attribute information and a preset motion logic of the target object; a state determination module, which is used to determine whether the target object is in a stationary state based on the state information.

[0115] Example 3

[0116] A person of ordinary skill in the art will appreciate that all or part of the steps in the various methods of the above embodiments may be completed by instructions, or by controlling related hardware through instructions. The instructions may be stored in a computer-readable storage medium and loaded and executed by a processor.

[0117] To this end, the third aspect of the present invention provides a storage medium, which is a computer-readable medium and stores a computer program. When the computer program is executed by a processor, the method provided by an embodiment of the present invention includes the following steps S11 to S12.

[0118] S11, during the process of automatic driving of the vehicle, matching object information of a target object within a preset range from the vehicle to obtain attribute information corresponding to the object information of the target object.

[0119] S12, determine the state information of the target object at least according to one of the first preset method, the second preset method, and the third preset method, the state information including at least one of the following: first state information, second state information, and third state information; wherein, the first preset method includes: inputting the attribute information of the target object into a data classification model, and determining the output result of the data classification model as the first state information, the second preset method includes: determining the second state information according to the attribute information of the target object and a preset convergence confidence threshold of the target object; the third preset method includes: determining the third state information according to the attribute information and a preset motion logic of the target object.

[0120] S13: Determine whether the target object is in a stationary state based on the state information.

[0121] Furthermore, when the computer program is executed by a processor, other methods provided by any of the above embodiments of the present invention are implemented.

[0122] The embodiment of the present invention determines the corresponding coordinated control mode and coordinated value in response to the intervention request, and determines the control instructions of the engine and the motor based on the coordinated control mode and the coordinated value, fully considering the dynamic torque requirements that may appear in each control system during the operation of the hybrid vehicle, thereby realizing the coordinated distribution and output control of the dynamic torque, making the torque control more efficient and reliable, ensuring the driving safety of the vehicle, improving the driving comfort and power, and improving the economy level of the hybrid vehicle.

[0123] Example 4

[0124] In a fourth aspect, a fourth embodiment of the present invention provides an electronic device, such as Figure 6As shown, the electronic device at least includes a memory 501 and a processor 502. The memory 501 stores a computer program. The processor 502 implements the method provided by any embodiment of the present invention when executing the computer program on the memory 501. Exemplarily, the method executed by the electronic device computer program is as follows:

[0125] S21, during the process of automatic driving of the vehicle, matching object information of a target object within a preset range from the vehicle to obtain attribute information corresponding to the object information of the target object.

[0126] S22, determine the state information of the target object at least according to one of the first preset method, the second preset method, and the third preset method, the state information including at least one of the following: first state information, second state information, and third state information; wherein, the first preset method includes: inputting the attribute information of the target object into a data classification model, and determining the output result of the data classification model as the first state information, the second preset method includes: determining the second state information according to the attribute information of the target object and a preset convergence confidence threshold of the target object; the third preset method includes: determining the third state information according to the attribute information and a preset motion logic of the target object.

[0127] S23: Determine whether the target object is in a stationary state based on the state information.

[0128] In specific implementation, the above-mentioned mode determination module and torque determination module are stored in the memory 501 as program units, and the processor 502 executes the above-mentioned program units stored in the memory 501 to implement corresponding functions.

[0129] The motion state determination method, device and storage medium described in the present invention determine the corresponding coordinated control mode and coordination value in response to the intervention request, and determine the control instructions of the engine and the motor based on the coordinated control mode and coordination value, fully considering the dynamic torque intervention requirements of each control system during the operation of the hybrid vehicle, thereby realizing the coordinated distribution and output control of the dynamic torque, making the control more efficient and reliable, ensuring the driving safety of the vehicle, improving the driving comfort and power, and improving the economy level of the hybrid vehicle.

[0130] The storage medium may be included in the electronic device, or may exist independently without being installed in the electronic device.

[0131] The storage medium carries one or more programs. When the one or more programs are executed by the electronic device, the electronic device: obtains at least two Internet Protocol addresses; sends a node evaluation request including at least two Internet Protocol addresses to a node evaluation device, wherein the node evaluation device selects an Internet Protocol address from the at least two Internet Protocol addresses and returns it; receives the Internet Protocol address returned by the node evaluation device; wherein the obtained Internet Protocol address indicates an edge node in a content distribution network.

[0132] Alternatively, the storage medium carries one or more programs, and when the one or more programs are executed by the electronic device, the electronic device: receives a node evaluation request including at least two Internet Protocol addresses; selects an Internet Protocol address from the at least two Internet Protocol addresses; and returns the selected Internet Protocol address; wherein the received Internet Protocol address indicates an edge node in a content distribution network.

[0133] Computer program code for performing the operations of the present invention may be written in one or more programming languages ​​or a combination thereof, including, but not limited to, object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the passenger computer, partially on the passenger computer, as a separate software package, partially on the passenger computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the passenger computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0134] It should be noted that the storage medium of the present invention can be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media can include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present invention, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present invention, a computer-readable signal medium can include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal can take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The computer readable signal medium may also be any storage medium other than a computer readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the storage medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0135] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present invention. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0136] The units involved in the embodiments of the present invention may be implemented by software or hardware, wherein the name of a unit does not limit the unit itself in some cases.

[0137] The functions described above herein may be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), and the like.

[0138] In the context of the present invention, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0139] The above description is only a preferred embodiment of the present invention and an explanation of the technical principles used. Those skilled in the art should understand that the disclosure scope involved in the present invention is not limited to the technical solution formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above disclosed concept. For example, the above features are replaced with the technical features with similar functions disclosed in the present invention (but not limited to) by each other.

[0140] In addition, although each operation is described in a specific order, this should not be construed as requiring these operations to be performed in the specific order shown or in a sequential order. Under certain circumstances, multitasking and parallel processing may be advantageous. Similarly, although some specific implementation details are included in the above discussion, these should not be construed as limiting the scope of the present invention. Some features described in the context of a separate embodiment can also be implemented in a single embodiment in combination. On the contrary, the various features described in the context of a single embodiment can also be implemented in multiple embodiments individually or in any suitable sub-combination mode.

[0141] Although the subject matter has been described in language specific to structural features and / or methodological logical actions, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or actions described above. On the contrary, the specific features and actions described above are merely example forms of implementing the claims.

[0142] A number of embodiments of the present invention have been described in detail above, but the present invention is not limited to these specific embodiments. Those skilled in the art can make various variations and modifications to the embodiments based on the concept of the present invention, and these variations and modifications should all fall within the scope of protection claimed by the present invention.

Claims

1. A method for determining a motion state, characterized in that: The method comprises: During the process of automatic driving of the vehicle, object information of a target object within a preset range from the vehicle is matched to obtain attribute information corresponding to the object information of the target object; The state information of the target object is determined at least in accordance with one of a first preset method, a second preset method, and a third preset method, wherein the state information includes at least one of the following: first state information, second state information, and third state information; wherein the first preset method includes: inputting the attribute information of the target object into a data classification model, and determining the output result of the data classification model as the first state information, and the second preset method includes: determining the second state information according to the attribute information of the target object and a preset convergence confidence threshold of the target object; and the third preset method includes: determining the third state information according to the attribute information and a preset motion logic of the target object; determining whether the target object is in a stationary state based on the state information; Matching the object information of the target object within a preset range from the vehicle to obtain attribute information corresponding to the object information of the target object, including: obtaining a first position value of the target object in the object information and a first size value of the target object in the object information; searching for first feature information matching the object information from a pre-stored feature information table, wherein the first feature information includes a second position value of the target object and a second size value of the target object, and the difference between the first position value and the second position value is less than a first threshold value and / or the difference between the first size value and the second size value is less than a second threshold value; obtaining second feature information corresponding to the first feature information from the feature information table, the second feature information including at least one of the following: speed information of the target object, direction information of the target object, variance information of the target object, association confidence information of the target object, and detection confidence information of the target object; determining the second feature information as the attribute information corresponding to the object information of the target object; The process of determining the first state information of the target object based on the first preset method includes: inputting the speed information of the target object, the direction information of the target object, the variance information of the target object, the association confidence information of the target object, and the detection confidence information of the target object in the attribute information of the target object into the data classification model at the same time, and the data classification model is used to determine the state information of the target object; determining the output result of the data classification model as the first state information of the target object, wherein the first state information includes one of the following: action state, static state; The process of determining the second state information of the target object based on the second preset method includes: parsing the detection speed and the filtered speed of the target object from the speed information of the target object, the filtered speed represents the speed after the detection speed of the target object is subjected to noise filtering; calculating the speed noise value of the target object according to the detection speed of the target object and the filtered speed of the target object; determining the convergence confidence corresponding to the speed information of the target object according to the speed noise value; determining the second state information of the target object according to the convergence confidence and the preset convergence confidence threshold, wherein the second state information of the target object includes one of the following: a speed value convergence state, a speed value non-convergence state; determining the second state information of the target object according to the convergence confidence and the preset convergence confidence threshold includes: when the convergence confidence is greater than the preset convergence confidence threshold, determining the second state information of the target object as a speed value convergence state; when the convergence confidence is not greater than the preset convergence confidence threshold, determining the second state information of the target object as a speed value non-convergence state; The process of determining the third state information of the target object based on the third preset mode includes: determining the movement direction of the target object within a preset time period according to the attribute information of the target object; in the case where the movement direction is determined to be unidirectional, determining multiple groups of movement time periods from the preset time period, each group of movement time periods in the multiple groups of movement time periods includes a first time period and a second time period, the end time of the first time period in each group is the start time of the second time period, and the duration of the first time period is the same as the duration of the second time period; obtaining a first moving distance of the target object in the first time period of each group of movement time periods, a second moving distance of the target object in the second time period of each group of movement time periods, and determining a first number of movement time periods in which the first moving distance is greater than the second moving distance, and a second number of movement time periods in which the first moving distance is not greater than the second moving distance; determining the third state information of the target object according to the first number and the second number, wherein determining the third state information of the target object according to the first number and the second number includes: in the case where the first number is less than the second number, determining that the third state information of the target object is a stationary state; in the case where the first number is not less than the second number, determining that the third state information of the target object is a moving state.

2. The determination method according to claim 1, characterized in that: Determining a convergence confidence level corresponding to the speed information of the target object according to the speed noise value includes: Calculating a first speed convergence value corresponding to the speed information of the target object at a first moment according to a preset convergence noise threshold and the speed noise value, wherein the first moment represents an initial moment; The first velocity convergence value is determined as a first convergence confidence level of the velocity information of the target object at the first moment.

3. The determination method according to claim 1, characterized in that: Determining a convergence confidence level corresponding to the speed information of the target object according to the speed noise value includes: Calculating a historical speed noise value of the target object according to a historical detection speed of the target object and a filtered speed of the target object; Calculate the historical speed convergence value corresponding to the historical speed information of the target object based on a preset convergence noise threshold and the historical speed noise value; Setting the second state information of the target object at the first moment as the second state information of the target object at the second moment, wherein the second moment is after the first moment; When the second state information of the target object at the second moment is a velocity value convergence state, determining an average value of the historical velocity convergence values ​​as a second convergence confidence level of the velocity information of the target object at the second moment; When the second state information of the target object at the second moment is a speed value non-convergence state, the minimum value of the historical speed convergence values ​​is determined as the second convergence confidence of the speed information of the target object at the second moment.

4. The determination method according to claim 1, characterized in that: Determining whether the target object is in a stationary state based on the state information includes: Determining the motion state of the target object based on the first state information, the second state information, and the third state information includes: In a case where it is determined that the first state information is a stationary state, the second state information is a velocity value convergence state, and the third state information is a stationary state, determining that the state information of the target object is a stationary state; In the case that any one of determining that the first state information is a stationary state, the second state information is a velocity value convergence state, and the third state information is a stationary state is not established, the state information of the target object is determined to be a moving state.

5. A device for determining a motion state, characterized in that: The device executes the method for determining the motion state as described in any one of claims 1 to 4.

6. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored program, wherein when the program is executed by a processor, the method for determining the motion state according to any one of claims 1 to 4 is implemented.

Images