Vehicle control method and device based on side parking of preceding vehicle, vehicle and medium

Abstract

The application discloses a vehicle control method and device based on side parking of a front vehicle, a vehicle and a medium. The method comprises the following steps: when a host vehicle is driving on a road with a queue of illegally parked vehicles and the host vehicle is driving on a lane beside the queue of illegally parked vehicles, it is judged whether there is a vacant parking space in the queue of illegally parked vehicles; if there is the vacant parking space in the queue of illegally parked vehicles, it is judged whether the front vehicle has a side parking intention of parking in the parking space; if the front vehicle has the side parking intention, a side parking track of the front vehicle is predicted, and the host vehicle is controlled to avoid driving according to the side parking track. According to the application, it can be known in advance that the front vehicle is ready for side parking by judging whether the front vehicle has the side parking intention, and then the host vehicle can avoid driving in time according to the side parking track of the front vehicle, so that the host vehicle does not need to react until the front vehicle starts to reverse, the reversing route of the front vehicle is avoided, the side parking behavior of the front vehicle can be recognized earlier, and the traffic efficiency of the vehicle is improved.

Description

TECHNICAL FIELD

[0001] Embodiments of the present application relate to the technical field of automatic driving, and in particular to a vehicle control method and device based on side parking of a preceding vehicle, a vehicle and a medium. BACKGROUND

[0002] In the process of driving from a starting point to a destination, an automatic driving vehicle needs to not only avoid various obstacles on the road to ensure safety, but also cope with various special operation scenarios. These scenarios include, but are not limited to, identifying and coping with various abnormal behaviors of a preceding vehicle, so as to timely escape when problems are encountered. In some specific situations, for example, when passing some road sections with more illegal parking on the right side, the automatic driving vehicle may encounter a situation where a social vehicle is side parking in front. In this case, the automatic driving vehicle needs to yield in advance or timely detour to avoid collision or other dangerous situations. However, the existing automatic driving logic has some problems. For example, the automatic driving vehicle will predict the behavior of the preceding vehicle, and if it finds that the preceding vehicle is reversing, the host vehicle will stop at a safe distance. However, this logic has the following problems: first, the automatic driving vehicle needs to be able to timely identify that the preceding vehicle is reversing. If the identification is too late, or the host vehicle starts reversing too late, the host vehicle will be stuck behind the preceding vehicle, which not only affects the side parking of the preceding vehicle, but also causes trouble to the driving of the host vehicle. Second, if the preceding vehicle is in a reversing state in the current lane of the host vehicle, the host vehicle will usually wait for the preceding vehicle to complete the operation. However, this approach will seriously affect the traffic efficiency, because the host vehicle needs to wait for the preceding vehicle to complete the entire process of reversing and side parking. SUMMARY

[0003] Embodiments of the present application provide a vehicle control method and device based on side parking of a preceding vehicle, a vehicle and a medium, aiming to solve the problem that the existing automatic driving strategy causes the host vehicle to be stuck behind the preceding vehicle or the host vehicle to wait for the preceding vehicle to complete side parking, thereby affecting traffic efficiency.

[0004] In a first aspect, embodiments of the present application provide a vehicle control method based on side parking of a preceding vehicle, which comprises:

[0005] When the ego vehicle is driving on a road with a queue of illegal parking and the ego vehicle is driving in a lane beside the queue of illegal parking, it is determined whether there is a vacant parking space in the queue of illegal parking;

[0006] If there is a vacant parking space in the queue of illegal parking, it is determined whether the preceding vehicle has a side parking intention of parking in the parking space;

[0007] If the preceding vehicle has the side parking intention, the side parking trajectory of the preceding vehicle is predicted, and the ego vehicle is controlled to avoid driving according to the side parking trajectory.

[0008] In a second aspect, the embodiments of the present application further provide a vehicle control device based on side parking of a front vehicle, comprising a unit for executing the method described above.

[0009] In a third aspect, the embodiments of the present application further provide a vehicle, comprising a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the method described above.

[0010] In a fourth aspect, the embodiments of the present application further provide a computer readable storage medium, which stores a computer program, and the computer program, when executed by a processor, can implement the method described above.

[0011] The embodiments of the present application provide a vehicle control method, device, vehicle and medium based on side parking of a front vehicle. The method comprises: when a host vehicle is driving on a road with a queue of illegal parking and the host vehicle is driving on a lane beside the queue of illegal parking, determining whether there is a vacant parking space in the queue of illegal parking; if there is a vacant parking space in the queue of illegal parking, determining whether the front vehicle has a side parking intention of parking in the parking space; if the front vehicle has the side parking intention, predicting a side parking track of the front vehicle, and controlling the host vehicle to avoid driving according to the side parking track. The present application can know in advance that the front vehicle is preparing to park on the side by determining whether the front vehicle has a side parking intention, and then make an avoiding driving according to the side parking track of the front vehicle in time, without making a response until the front vehicle starts to reverse, avoiding blocking the reversing route of the front vehicle, and being able to identify the side parking behavior of the front vehicle earlier, thereby improving the traffic efficiency of the vehicle. BRIEF DESCRIPTION OF DRAWINGS

[0012] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without creative labor.

[0013] Figure 1 The step flowchart of the vehicle control method based on side parking of a front vehicle of the embodiments of the present application;

[0014] Figure 2 The sub-step schematic diagram of the vehicle control method based on side parking of a front vehicle of the embodiments of the present application;

[0015] Figure 3 The schematic diagram of the distribution type of the embodiments of the present application;

[0016] Figure 4A schematic diagram of a longitudinal distribution of different parking modes of embodiments of the present application;

[0017] Figure 5 A schematic diagram of a step flow of a vehicle control method based on side parking of a preceding vehicle of another embodiment of the present application;

[0018] Figure 6 A schematic diagram of a step flow of a vehicle control method based on side parking of a preceding vehicle of yet another embodiment of the present application;

[0019] Figure 7 A schematic diagram of a step flow of a vehicle control method based on side parking of a preceding vehicle of still another embodiment of the present application;

[0020] Figure 8 A schematic block diagram of a vehicle control device based on side parking of a preceding vehicle provided by embodiments of the present application;

[0021] Figure 9 A schematic block diagram of a vehicle provided by embodiments of the present application. DETAILED DESCRIPTION

[0022] The technical solutions in the embodiments of the present application will be clearly and completely described with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some but not all of the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative work fall within the protection scope of the present application.

[0023] It should be understood that the terms "comprising" and "including" as used in the specification and the appended claims indicate the presence of the described features, integers, steps, operations, elements, and / or components but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof.

[0024] It should also be understood that the terms used in the present application specification are only for the purpose of describing particular embodiments and are not intended to limit the present application. As used in the present application specification and the appended claims, the singular forms "a", "an" and "the" are intended to include the plural forms unless the context clearly indicates otherwise.

[0025] It should be further understood that the term "and / or" as used in the present application specification and the appended claims means one or more of the associated listed items as well as all possible combinations of the items and includes the combinations.

[0026] As used in the specification and the appended claims, the term “if’ can be interpreted as meaning “when,” or “upon,” or “in response to a determination,” or “in response to a detection” depending on the context. Similarly, the phrase “if it is determined” or “if [the described condition or event] is detected” can be interpreted as meaning “upon a determination” or “in response to a determination” or “upon detecting [the described condition or event]” or “in response to detecting [the described condition or event],” depending on the context.

[0027] Referring to Figure 1 , Figure 1 A flowchart of a vehicle control method based on side parking of a preceding vehicle is provided for an embodiment of the present application. The vehicle control method based on side parking of a preceding vehicle is applied in an autonomous vehicle. The vehicle control method based on side parking of a preceding vehicle is described in detail as follows. As shown in Figure 1 The method comprises the following steps: S110-S130.

[0028] S110, when the ego vehicle is driving on a road with a queue of illegally parked vehicles and the ego vehicle is driving on a lane beside the queue of illegally parked vehicles, determining whether there is a vacant parking space in the queue of illegally parked vehicles.

[0029] In this embodiment, the queue of illegally parked vehicles refers to a queue of multiple vehicles parked on the roadside in violation of regulations. The queue of illegally parked vehicles can be a continuous queue of multiple vehicles parked or a non-continuous queue with vacant parking spaces in between. The queue of illegally parked vehicles is likely to occur on busy roads, residential areas, commercial areas, and temporary construction areas, etc. The vehicle can identify the queue of illegally parked vehicles on the road through visual sensors, laser radars, millimeter wave radars, high-precision maps, vehicle-to-everything (V2X) communication, machine learning algorithms, etc. The vehicle can identify the queue of illegally parked vehicles in any one or more of the above-mentioned ways, which is not limited herein. It should be noted that the determination of the vacant parking space is only performed when the vehicle is driving on a road with a queue of illegally parked vehicles and simultaneously driving on a lane beside the queue of illegally parked vehicles. For example, the queue of illegally parked vehicles is located in the rightmost lane, and the determination of the vacant parking space is performed when the ego vehicle is driving on the second right lane. If the ego vehicle is driving on the third right lane or further outward, the determination of the vacant parking space is not necessary, because the vehicle usually performs side parking on the second right lane. When the vehicle meets the conditions of driving on a road with a queue of illegally parked vehicles and simultaneously driving on a lane beside the queue of illegally parked vehicles, the determination of the vacant parking space in the queue of illegally parked vehicles is started.

[0030] The judgment of the vacant parking space in the illegal parking queue adopts the following ways. One is to recognize by visual sensor. The high-definition camera can capture the image of the road surface. The parking line and the contour of the vehicle on the road surface are recognized by image processing and computer vision algorithm. When the distance between the adjacent vehicles is greater than the length of the normal vehicle plus the safety distance, the system can determine that there is a vacant parking space in this area. The second is to recognize by laser radar. The laser radar can generate high-precision three-dimensional point cloud map to determine the three-dimensional geometric structure of the surrounding environment. By analyzing the point cloud data, the autonomous vehicle can accurately determine the distance and gap between each vehicle, and further identify the space available for parking. The third is to recognize by ultrasonic sensor and millimeter wave radar. Ultrasonic sensor and millimeter wave radar are commonly used for close-range detection. They can help the vehicle accurately measure the distance to the surrounding obstacles (including vehicles), especially during low-speed driving and parking stages. These sensors help to identify small gaps between illegal parking vehicles. The fourth is to recognize by high-precision map. The autonomous vehicle usually uses a high-precision map which contains the exact location and size information of each parking space. By combining with real-time sensor data, the vehicle can locate itself on the map and find the vacant parking space. Through one or more of the above ways, it can be accurately determined whether there is a vacant parking space in the illegal parking queue. The present embodiment is not limited in this regard.

[0031] S120, if there is a vacant parking space in the illegal parking queue, determining whether the front vehicle has a side parking intention of parking in the parking space.

[0032] In the present embodiment, the side parking intention refers to the premonition of the vehicle for side parking. Specifically, for example, it can be to slow down in front of the side parking space gap, or to twist the direction to the side of the side parking space gap, or other ways, which depends on the parking method of side parking and the shape and size of the parking space. For example, the parking space is a vertical gap, and the normal side parking method is adopted. The vehicle first drives straight to the position beside the front vehicle in front of the parking space, and then turns the direction to reverse into the garage. In this scenario, the vehicle slowly drives to the position beside the front vehicle in front of the parking space, which is considered to have a side parking intention. For another example, the parking space is a vertical gap, and the first head-in side parking method is adopted. The vehicle first twists the direction to the side of the gap, and then twists to the left side of the gap to the front of the gap, and finally turns the direction to reverse into the garage. In this scenario, the vehicle twists the direction to the side of the gap, which is considered to have a side parking intention. Of course, it can be understood that there can be other judgment methods, which are not limited herein.

[0033] In an embodiment, as shown in Figure 2 S120, the step S120 includes S121-S122.

[0034] S121, identifying whether the front vehicle performs a one-stage side parking behavior of parking into the parking space by a pre-trained side parking model, wherein the one-stage side parking behavior is a vehicle adjustment behavior before a vehicle enters a garage;

[0035] S122, if the front vehicle performs the one-stage side parking behavior of parking into the parking space, determining that the front vehicle has a side parking intention of parking into the parking space.

[0036] In the embodiment, the behavior of a general side parking can be divided into two stages. The one-stage side parking behavior is a vehicle adjustment behavior before a vehicle enters a garage, mainly adjusting the position of the vehicle, and placing the position of the vehicle to facilitate the vehicle to enter the garage. For example, the vehicle slowly stops in front of a side parking space gap and turns a direction to the side of the side parking space gap. The two-stage side parking behavior is a vehicle entering a garage behavior, that is, the vehicle slowly reverses into the side parking space at an inclined angle. The behavior of the two-stage side parking behavior may, for example, be turning on a reversing light or reversing. The existing side parking recognition method usually starts to recognize at the second stage, that is, it is known that the front vehicle wants to side park when the front vehicle reverses. Therefore, in order to recognize the side parking intention of the front vehicle earlier and more accurately, the embodiment determines that the front vehicle has a side parking intention by recognizing the one-stage side parking behavior of the front vehicle. Specifically, a side parking model is used to recognize the one-stage side parking behavior of the front vehicle. The side parking model is pre-trained. The side parking model can be trained by using a neural network model, a decision tree model, a K-nearest neighbor model, etc. Regardless of the algorithm model, as long as it can recognize the one-stage side parking behavior of the front vehicle, it is not limited herein. When the side parking model recognizes the one-stage side parking behavior of the front vehicle, it is determined that the front vehicle has a side parking intention, so that the ego vehicle can make a response decision in time.

[0037] In the embodiment, the training steps of the side parking model include: obtaining vehicle data of a vehicle driving near a side parking space on a roadside, obtaining a placement distribution type of a queue of illegal parking on the roadside, and combining the vehicle data and the placement distribution type to form a training data, wherein the vehicle data includes a vehicle position and a vehicle speed corresponding to each frame in a driving process, and the placement distribution type includes a longitudinal arrangement, a transverse arrangement, and an oblique arrangement; performing positive and negative example labeling on the training data, wherein the training data with negative speed is labeled as a positive example, and the training data without negative speed is labeled as a negative example; and inputting the labeled training data into a deep learning model to train the side parking model, wherein the side parking model is used to identify a one-stage side parking behavior of a vehicle when the vehicle faces the placement distribution type of the longitudinal arrangement, the transverse arrangement, and the oblique arrangement.

[0038] Specifically, since the identification and judgment of side parking depends on the parking method of side parking and the shape and size of the parking space, the identification of side parking behavior has a strong correlation with the parking method of side parking and the shape and size of the parking space, so a vehicle data (which can represent the parking method) near the side parking space and the placement distribution type of the illegal parking queue (which can represent the shape of the parking space gap) are used to form a training data, and a plurality of training data are obtained for model training.

[0039] It should be noted that the shape and size of the parking space are related to the placement distribution type of the illegal parking queue. Generally, as shown in Figure 3 , the placement distribution type of the illegal parking queue usually includes longitudinal arrangement, transverse arrangement and diagonal arrangement. The longitudinal arrangement is that the illegal parking vehicles are arranged in a vertical parking manner on the roadside, and the parking space gap of the longitudinal arrangement is a vertical gap; the transverse arrangement is that the illegal parking vehicles are arranged in a horizontal parking manner on the roadside, and the parking space gap of the transverse arrangement is a horizontal gap; the diagonal arrangement is that the illegal parking vehicles are arranged in an inclined angle parking manner on the roadside, which can be inclined upward or inclined downward, and the parking space gap of the diagonal arrangement is an inclined gap. Therefore, the shape of the parking space gap can be judged by the placement distribution type, and different side parking methods can be analyzed according to the shape of the parking space gap.

[0040] For different shapes of side parking space gaps, vehicles usually adopt different side parking methods, and for the same shape of side parking space gap, a plurality of different side parking methods can be adopted. For example, as shown in Figure 4 , for the vertical parking space gap of the longitudinal arrangement, the side parking method adopted by the vehicle usually includes the following four kinds, the first is the head-in parking method, the second is the tail-in parking method, the third is the normal side parking method, and the fourth is the reverse side parking method, and the driving track of each parking method is represented by the arrow in the figure, the first arrow represents the first stage of side parking, and the second arrow represents the second stage of side parking. Of course, it can be understood that for other placement distribution types such as transverse arrangement and diagonal arrangement, each corresponds to a plurality of different side parking methods, which will not be illustrated one by one here. Therefore, by collecting the data of the vehicle adopting different side parking methods for side parking under different placement distribution types as training data for model training, an accurate identification of side parking behavior in various scenarios can be trained.

[0041] The vehicle data includes the vehicle position and vehicle speed corresponding to each frame during the driving of the vehicle near the side parking space. Through the vehicle position and vehicle speed data from the first frame to the last frame, the driving track of the vehicle near the side parking space can be clearly presented, so that the side parking manner of the vehicle can be analyzed. In other embodiments, in order to further improve the recognition accuracy of the model, the vehicle data such as the vehicle type (car, van, bus), acceleration and yaw angle of the vehicle can also be obtained to form the training data for training.

[0042] Before training, the training data needs to be labeled first. Since part of the obtained training data is that the vehicle has parked in the side parking space, and part of the data is that the vehicle has not parked in the side parking space. The training data in which the vehicle has parked in the side parking space is labeled as a positive example, and the training data in which the vehicle has not parked in the side parking space is labeled as a negative example. Specifically, whether the training data has parked in the side parking space can be known by analyzing the vehicle speed. Generally, the vehicle needs to perform reverse parking when parking in the side parking space, and the vehicle speed corresponding to the reverse parking is negative speed, that is, negative value. Therefore, by judging whether there is a continuous negative speed in the vehicle speed of multiple frames, it can be confirmed that the vehicle of the training data has parked in the side parking space, and therefore labeled as a positive example. On the contrary, without negative speed, it indicates that the vehicle has not parked in the side parking space, and the training data is labeled as a negative example.

[0043] After labeling, first, feature engineering is performed on the training data, and the training data is converted into features suitable for model input, and then a module is constructed, which can be a convolutional neural network model, a recurrent neural network module, a spatio-temporal convolution network, etc. The labeled training data is divided into a training set, a validation set and a test set. Then, using the back propagation algorithm and other optimization strategies (such as Adam, SGD, etc.), the neural network model is trained by the positive and negative example data in the training set, the goal of which is to enable the model to distinguish between one-stage side parking behavior and non-side parking behavior. Then, a suitable loss function, such as cross-entropy loss, is designed to measure the gap between the model prediction result and the actual label, and evaluation indicators such as accuracy, recall rate and F1 score are set to monitor the performance of the model on the validation set. Then, the model parameters are adjusted according to the performance on the validation set, including but not limited to learning rate, network layer number, node number, etc. Then, the model performance is evaluated on an independent test set to ensure that the model not only performs well on the training set, but also effectively identifies the side parking behavior on new data. Finally, after sufficient verification, the optimal model is fine-tuned or integrated to enable it to accurately identify the side parking behavior of various distributions in real-time applications. In summary, the side parking model of the embodiment of the present application can accurately identify the one-stage side parking behavior of the preceding vehicle, greatly improving the accuracy and timeliness of the identification.

[0044] S130, if the front vehicle has the side parking intention, predicting a side parking trajectory of the front vehicle, and controlling the ego vehicle to avoid driving according to the side parking trajectory.

[0045] In this embodiment, after it is judged that the front vehicle has the side parking intention, the ego vehicle immediately predicts a side parking trajectory of the front vehicle, and the ego vehicle considers that the front vehicle will park sideways according to the predicted side parking trajectory, so that the ego vehicle makes timely avoiding driving behavior based on the predicted side parking trajectory, such as stopping, reversing, detouring, reversing and detouring, and allowing vehicles to detour, etc. Among them, the prediction of the side parking trajectory can take the following ways, one is to use multiple sensors such as cameras, laser radars, millimeter wave radars, ultrasonic sensors, etc., to continuously collect real-time position, speed, acceleration, direction angle and other motion state information of the front vehicle, as well as environmental information around the front vehicle, including road structure, parking space identification, pedestrians and other vehicles, etc. to predict the side parking trajectory of the front vehicle; the second is to use space-time sequence prediction model such as Social LSTM, Intent Net, GAT-LSTM, etc. These models can combine historical trajectory data and environmental information to predict future trajectory, combine temporal convolutional neural network (TCNs) or long short-term memory network (LSTMs) and other recurrent neural network structures to establish a dynamic model of the front vehicle, and consider the physical constraints and driving behavior logic of the vehicle to predict the side parking trajectory of the front vehicle; the third is to use high-precision map information to clearly define the road layout, including the specific position, size and direction of the side parking space, so as to analyze the positional relationship of the front vehicle relative to the roadside parking space, and the relative position of the front vehicle with other vehicles on the roadside, and calculate the side parking trajectory of the front vehicle based on this. Of course, it can be understood that other trajectory prediction methods can also be used as long as the future side parking trajectory of the front vehicle can be predicted, which is not limited here.

[0046] In an embodiment, as shown in FIG. 13B, the step S130 includes steps S231-S237. Figure 5

[0047] S231, if the front vehicle has the side parking intention and the ego vehicle is following the front vehicle, controlling the ego vehicle to slow down and keep a safe distance from the front vehicle until the front vehicle stops;

[0048] S232, judging whether the front vehicle performs a two-stage side parking behavior, wherein the two-stage side parking behavior is a vehicle entering a garage behavior;

[0049] S233, if the front vehicle performs a two-stage side parking behavior, predicting a side parking trajectory of the front vehicle, and judging whether there is enough detour space for the ego vehicle to detour the front vehicle in front of and on the side of the ego vehicle;

[0050] ​S234, if the bypass space exists, controlling the ego vehicle to bypass or reverse bypass according to the side parking trajectory;

[0051] S235, if the bypass space does not exist, controlling the ego vehicle to stop or reverse to give way according to the side parking trajectory.

[0052] In an embodiment, after it is determined that the front vehicle has a side parking intention, since the ego vehicle and the front vehicle are in a following driving state at this time, the distance between the front and back of the two vehicles is relatively close, the speed of the ego vehicle is first controlled to slow down, and then the front vehicle is waited until it stops. At this time, it is necessary to observe whether the front vehicle is truly intended to perform side parking, that is, to make a judgment on the two-stage side parking behavior. Generally, the two-stage side parking behavior includes turning on the reverse light of the front vehicle or observing the trend of reversing. If the above behaviors occur, the side parking trajectory of the front vehicle is predicted, and it is determined whether there is enough space in front of and on the side of the ego vehicle for the ego vehicle to bypass. The space in front is the space distance between the ego vehicle and the front vehicle, and the space on the side is the space distance on the left side of the ego vehicle. If the bypass space is sufficient, the ego vehicle is controlled to bypass or reverse bypass based on the predicted side parking trajectory. For example, if the bypass trajectory of the ego vehicle does not overlap with the predicted side parking trajectory, the ego vehicle is directly controlled to bypass; if the bypass trajectory of the ego vehicle overlaps with the predicted side parking trajectory, it means that the space is not enough, and the ego vehicle is controlled to reverse bypass, that is, to reverse to create space and then bypass. If the bypass space is not enough, the ego vehicle is controlled to stop or reverse to give way based on the predicted side parking trajectory. For example, if the ego vehicle is outside the side parking trajectory, the ego vehicle is controlled to stop and wait for the front vehicle to complete side parking before passing; if the ego vehicle is on the side parking trajectory, it means that the front vehicle will collide with the ego vehicle when reversing, and the ego vehicle is controlled to reverse to avoid the front vehicle and wait for the front vehicle to complete side parking before passing. Thus, in the following driving scenario, different driving decisions are made based on different predicted trajectories, ensuring safe vehicle passing and improving passing efficiency.

[0053] S236, if the front vehicle is identified as a parking violation vehicle, controlling the ego vehicle to bypass according to the current traffic condition;

[0054] S237, if the front vehicle is identified as a parking violation vehicle, controlling the ego vehicle to bypass according to the current traffic condition;

[0055] In the embodiment, for the road with many illegal parking vehicles, the illegal parking vehicles can also make other illegal parking manners. For example, directly parking in the lane on the left of the illegal parking queue to form two illegal parking queues, or in some cases, the front vehicle is waiting for passing due to the congestion in front. For these scenes, the ego vehicle can identify the front vehicle as an illegal parking vehicle, and based on the normal automatic driving logic, the front vehicle is regarded as an obstacle, and the ego vehicle is controlled to detour according to the current traffic condition. If the front vehicle does not respond for a period of time after stopping, the front vehicle is identified as a static vehicle, and the ego vehicle is controlled to detour or continue following the vehicle according to the current traffic condition.

[0056] In an embodiment, as shown in Figure 6 the step S130 comprises: S331-S333.

[0057] S331, if the front vehicle has the side parking intention and the ego vehicle is not following the front vehicle, predicting the side parking trajectory of the front vehicle, and judging whether there is enough detour space for the ego vehicle to detour the front vehicle on the side of the ego vehicle;

[0058] S332, if the detour space exists, controlling the ego vehicle to detour according to the side parking trajectory;

[0059] S333, if the detour space does not exist, controlling the ego vehicle to slow down and stop outside the side parking trajectory of the front vehicle, and when there is a detour space in front of and on the side of the ego vehicle, controlling the ego vehicle to detour.

[0060] In an embodiment, after judging that the front vehicle has the side parking intention, since the ego vehicle and the front vehicle are not following each other at this time, the distance between them is far, and the ego vehicle has enough time to deal with the side parking of the front vehicle, compared with the coping strategy of following the vehicle, the ego vehicle does not need to slow down and wait for the front vehicle to stop, and can directly predict the side parking trajectory of the front vehicle, and judge whether there is enough space for the ego vehicle to detour on the side of the ego vehicle. If the detour space is enough, the ego vehicle is controlled to detour based on the predicted side parking trajectory. If the detour space is not enough, the ego vehicle is first slowed down and stopped outside the side parking trajectory of the front vehicle to avoid blocking the front vehicle, and then the ego vehicle is controlled to detour when the front vehicle enters the garage or the front vehicle provides enough space. Thus, in the scene of not following the vehicle, different driving decisions are made based on different predicted trajectories, the safe passing of the vehicle is ensured, and the passing efficiency is improved.

[0061] In an embodiment, as shown in Figure 7 the vehicle control method based on the side parking of the front vehicle further comprises steps: S140-S160.

[0062] S140, judging whether the distance between the ego vehicle and the front vehicle and the distance between the ego vehicle and the rear vehicle exceed a safety distance threshold;

[0063] S150, if the distance between the ego vehicle and the front vehicle exceeds the safety distance threshold and the distance between the ego vehicle and the rear vehicle does not exceed the safety distance threshold, controlling the ego vehicle to back up;

[0064] S160, if the distance between the ego vehicle and the front vehicle and the distance between the ego vehicle and the rear vehicle both exceed the safety distance threshold, controlling the ego vehicle to honk.

[0065] In the embodiment, one safety distance threshold is set for each vehicle to determine whether the front and rear vehicles are too close. If the front vehicle backs up and intrudes the safety threshold distance of the ego vehicle, the ego vehicle needs to back up to avoid. Meanwhile, the rear vehicle needs to give the ego vehicle enough space to continue backing up on the premise of keeping a safety distance from the rear obstacle. If the front vehicle is too close to the ego vehicle and enters a dangerous distance, honking is needed.

[0066] Figure 8 is a schematic block diagram of a vehicle control device 300 based on side parking of a front vehicle provided by the embodiment. As shown in Figure 8 corresponding to the above vehicle control method based on side parking of a front vehicle, the embodiment also provides a vehicle control device 300 based on side parking of a front vehicle. The vehicle control device 300 based on side parking of a front vehicle includes units for executing the above vehicle control method based on side parking of a front vehicle, and the device can be configured in a vehicle. Specifically, please refer to Figure 8 , the vehicle control device 300 based on side parking of a front vehicle includes a parking space judgment unit 301, an intention judgment unit 302, and a control unit 303.

[0067] The parking space judgment unit 301 is configured to determine whether there is a vacant parking space in the illegal parking queue when the ego vehicle drives on a road with an illegal parking queue and drives on a lane beside the illegal parking queue. The intention judgment unit 302 is configured to determine whether the front vehicle has a side parking intention to park in the parking space if there is a vacant parking space in the illegal parking queue. The control unit 303 is configured to predict a side parking trajectory of the front vehicle and control the ego vehicle to avoid driving according to the side parking trajectory if the front vehicle has the side parking intention.

[0068] In an embodiment, the intention judgment unit 302 includes a stage recognition unit and an intention determination unit.

[0069] Among them, the first-stage recognition unit is used to identify whether the front vehicle performs a first-stage parallel parking behavior in the parking space through a pre-trained parallel parking model, wherein the first-stage parallel parking behavior is the vehicle adjustment behavior before the vehicle enters the parking space; the intention judgment unit is used to judge whether the front vehicle has the intention to park in the parking space if the front vehicle performs a first-stage parallel parking behavior in the parking space.

[0070] In one embodiment, the first-stage recognition unit includes: an acquisition unit, a labeling unit, and a training unit.

[0071] Among them, the acquisition unit is used to obtain vehicle data of vehicles traveling near the side parking spaces on the roadside, obtain the placement distribution type of the illegally parked queues on the roadside, and form the vehicle data and the placement distribution type into a training data, wherein the vehicle data includes the vehicle position and vehicle speed corresponding to each frame during the driving process, and the placement distribution type includes longitudinal arrangement, transverse arrangement and diagonal arrangement; the labeling unit is used to label the training data with positive and negative examples, wherein the training data with negative speed of the vehicle is labeled as positive example, and the training data without negative speed of the vehicle is labeled as negative example; the training unit is used to input the labeled training data into the deep learning model for training to obtain the side parking model, wherein the side parking model is used to identify the side parking behavior of the vehicle in a stage when facing the placement distribution types of the longitudinal arrangement, the transverse arrangement and the diagonal arrangement respectively.

[0072] In one embodiment, the control unit 303 includes: a deceleration unit, a two-stage judgment unit, a first prediction unit, a first detour unit, a yielding unit, a second detour unit, and a third detour unit.

[0073] The deceleration unit is configured to control the ego vehicle to decelerate and keep a safe distance from the front vehicle until the front vehicle stops if the front vehicle has the side parking intention and the ego vehicle is following the front vehicle; the two-stage judgment unit is configured to judge whether the front vehicle performs a two-stage side parking behavior, wherein the two-stage side parking behavior is a vehicle parking behavior; the first prediction unit is configured to predict a side parking trajectory of the front vehicle and judge whether a front side of the ego vehicle has a sufficient detour space for the ego vehicle to detour the front vehicle if the front vehicle performs the two-stage side parking behavior; the first detour unit is configured to control the ego vehicle to detour or reverse detour according to the side parking trajectory if the detour space exists; the yielding unit is configured to control the ego vehicle to stop or reverse yield according to the side parking trajectory if the detour space does not exist; the second detour unit is configured to control the ego vehicle to detour according to a current traffic condition if the front vehicle is identified as a parking violation vehicle; and the third detour unit is configured to control the ego vehicle to detour or control the ego vehicle to follow the front vehicle according to the current traffic condition if the front vehicle is identified as stopping and no driving behavior is detected within a preset time.

[0074] In an embodiment, the control unit 303 comprises a second prediction unit, a fourth detour unit and a fifth detour unit.

[0075] The second prediction unit is configured to predict a side parking trajectory of the front vehicle and judge whether a side of the ego vehicle has a sufficient detour space for the ego vehicle to detour the front vehicle if the front vehicle has the side parking intention and the ego vehicle is not following the front vehicle; the fourth detour unit is configured to control the ego vehicle to detour according to the side parking trajectory if the detour space exists; and the fifth detour unit is configured to control the ego vehicle to decelerate and stop outside the side parking trajectory and control the ego vehicle to detour when a front side and a side of the ego vehicle have a detour space if the detour space does not exist.

[0076] In an embodiment, the vehicle control device 300 based on the front vehicle side parking further comprises a distance judgment unit, a reverse unit and a horn unit.

[0077] The distance judgment unit is configured to judge whether distances between the ego vehicle and the front vehicle and the rear vehicle respectively exceed a safe distance threshold; the reverse unit is configured to control the ego vehicle to reverse if the distance between the ego vehicle and the front vehicle exceeds the safe distance threshold and the distance between the ego vehicle and the rear vehicle does not exceed the safe distance threshold; and the horn unit is configured to control the ego vehicle to sound a horn if the distances between the ego vehicle and the front vehicle and the rear vehicle respectively exceed the safe distance threshold.

[0078] The vehicle control device 300 based on the front vehicle side parking described above can be implemented in the form of a computer program which can run on a vehicle as shown in Figure 9 the vehicle.

[0079] See also Figure 9 , Figure 9 It is a schematic block diagram of a vehicle provided in an embodiment of the present application.

[0080] See Figure 9 The vehicle 500 includes a processor 502 , a memory, and a network interface 505 connected via a system bus 501 , wherein the memory may include a non-volatile storage medium 503 and an internal memory 504 .

[0081] The non-volatile storage medium 503 can store an operating system 5031 and a computer program 5032. The computer program 5032 includes program instructions, which, when executed, can enable the processor 502 to execute a vehicle control method based on parallel parking of a preceding vehicle.

[0082] The processor 502 is used to provide computing and control capabilities to support the operation of the entire vehicle 500.

[0083] The internal memory 504 provides an environment for the operation of the computer program 5032 in the non-volatile storage medium 503. When the computer program 5032 is executed by the processor 502, the processor 502 can execute a vehicle control method based on parallel parking of the preceding vehicle.

[0084] The network interface 505 is used to communicate with other devices through the network. Figure 9 The structure shown in the figure is merely a block diagram of a portion of the structure related to the solution of the present application, and does not constitute a limitation on the vehicle 500 to which the solution of the present application is applied. The specific vehicle 500 may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0085] The processor 502 is configured to run a computer program 5032 stored in the memory to implement the steps of the above method.

[0086] It should be understood that, in the embodiments of the present application, the processor 502 can be a central processing unit (CPU), and can also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor.

[0087] It can be understood by those skilled in the art that all or part of the processes in the above-mentioned embodiments can be completed by a computer program instructing related hardware. The computer program includes program instructions, and the computer program can be stored in a storage medium, which is a computer-readable storage medium. The program instructions are executed by at least one processor in the computer system to implement the process steps of the above-mentioned embodiments.

[0088] Therefore, the present application also provides a storage medium. The storage medium can be a computer-readable storage medium. The storage medium stores a computer program, wherein the computer program includes program instructions. The program instructions are executed by a processor to make the processor execute the steps of the above-mentioned method.

[0089] The storage medium can be a U disk, a mobile hard disk, a read-only memory (ROM), a magnetic disk or an optical disk, and various computer-readable storage media that can store program codes.

[0090] Those skilled in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be realized in electronic hardware, computer software or a combination of both. In order to clearly illustrate the interchangeability of hardware and software, the components and steps of each example have been described in the above description in general terms. Whether the functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. A person skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0091] In several embodiments provided by the present application, it should be understood that the disclosed apparatus and method can be implemented in other manners. For example, the described apparatus embodiments are merely schematic. For example, the division of the units is merely a logical function division. For example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In a possible implementation process, the steps of the described method can be performed in a different order, or can be omitted, or can be combined into another process, or some features can be ignored or not executed.

[0092] The steps in the method embodiments of the present application can be adjusted, combined and deleted according to actual needs. The units in the apparatus embodiments of the present application can be combined, divided and deleted according to actual needs. In addition, each functional unit in each embodiment of the present application can be integrated into a processing unit, or each unit can exist physically, or two or more units can be integrated into one unit.

[0093] The integrated unit, if realized in the form of a software functional unit and sold or used as an independent product, can be stored in a storage medium. Based on such understanding, the technical solutions of the present application essentially or the part that contributes to the prior art, or the whole or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, and includes a plurality of instructions for causing a vehicle to execute all or part of the steps of the method described in each embodiment of the present application.

[0094] In the above embodiments, the description of each embodiment is focused on, and the parts not described in detail in a certain embodiment can be referred to the relevant description of other embodiments.

[0095] Obviously, those skilled in the art can make various modifications and variations to the present application without departing from the spirit and scope of the present application. Thus, these modifications and variations also belong to the scope of the claims of the present application and their equivalent technologies, and the present application is intended to include these modifications and variations.

[0096] The above description is merely specific embodiments of the present application, but the protection scope of the present application is not limited thereto. Any skilled person in the art can easily think of various equivalent modifications or replacements within the technical range disclosed by the present application, and these modifications or replacements should be covered in the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A vehicle control method based on side parking of a preceding vehicle, characterized in that: The method comprises: When the vehicle is traveling on a road with an illegally parked vehicle queue and is traveling in a lane next to the illegally parked vehicle queue, determining whether there is a vacant parking space in the illegally parked vehicle queue; If there is a vacant parking space in the illegally parked queue, determining whether the preceding vehicle has a parallel parking intention to park in the parking space; If the preceding vehicle has the intention to park sideways and the self-vehicle is following the preceding vehicle, the self-vehicle is controlled to slow down and maintain a safe distance from the preceding vehicle until the preceding vehicle stops; it is determined whether the preceding vehicle performs a two-stage parallel parking behavior, wherein the two-stage parallel parking behavior is a vehicle parking behavior; if the preceding vehicle performs a two-stage parallel parking behavior, the parallel parking trajectory of the preceding vehicle is predicted, and it is determined whether there is sufficient space in front of and to the side of the self-vehicle for the self-vehicle to bypass the preceding vehicle; if there is such space, the self-vehicle is controlled to bypass or reverse and bypass according to the parallel parking trajectory; if there is no such space, the self-vehicle is controlled to brake or reverse and give way according to the parallel parking trajectory; If the preceding vehicle has the intention to park sideways and the own vehicle is not following the preceding vehicle, the lateral parking trajectory of the preceding vehicle is predicted, and it is determined whether there is sufficient detour space to the side of the own vehicle for the own vehicle to bypass the preceding vehicle; if the detour space exists, the own vehicle is controlled to bypass according to the lateral parking trajectory; if the detour space does not exist, the own vehicle is controlled to decelerate and brake to a stop outside the lateral parking trajectory, and the own vehicle is controlled to bypass when there is detour space in front of and to the side of the own vehicle.

2. The method according to claim 1, characterized in that The step of determining whether the preceding vehicle has a parallel parking intention to park in the parking space comprises: Identifying, using a pre-trained parallel parking model, whether the preceding vehicle has performed a phase of parallel parking behavior into the parking space, wherein the phase of parallel parking behavior is a vehicle adjustment behavior before the vehicle enters the parking space; If the preceding vehicle performs a phase of parallel parking in the parking space, it is determined that the preceding vehicle has an intention to parallel park in the parking space.

3. The method according to claim 2, characterized in that The training steps of the parallel parking model include: Obtaining vehicle data of vehicles traveling near a side parking space on the roadside, obtaining a placement distribution type of illegally parked vehicles on the roadside, and combining the vehicle data and the placement distribution type into a piece of training data, wherein the vehicle data includes the vehicle position and vehicle speed corresponding to each frame during driving, and the placement distribution type includes longitudinal arrangement, transverse arrangement, and diagonal arrangement; The training data is labeled as positive and negative examples, wherein the training data in which the vehicle speed has negative speed is labeled as a positive example, and the training data in which the vehicle speed does not have negative speed is labeled as a negative example; The labeled training data is input into a deep learning model for training to obtain the parallel parking model, wherein the parallel parking model is used to identify the parallel parking behavior of the vehicle in a stage when facing the placement distribution types of the longitudinal arrangement, the transverse arrangement and the diagonal arrangement.

4. The method according to claim 1, wherein After the step of controlling the vehicle to decelerate and maintain a safe distance from the preceding vehicle until the preceding vehicle stops, the method further includes: If the preceding vehicle is identified as illegally parked, the vehicle is controlled to detour according to the current traffic conditions; If it is identified that the preceding vehicle has stopped and has not moved within a preset time, the vehicle is controlled to detour or to follow the preceding vehicle according to the current traffic conditions.

5. The method according to claim 1, wherein The method further comprises: Determine whether the distance between the vehicle and the preceding and following vehicles exceeds the safety distance threshold; If the distance between the vehicle and the preceding vehicle exceeds the safety distance threshold and the distance between the vehicle and the following vehicle does not exceed the safety distance threshold, the vehicle is controlled to reverse. If the distances between the vehicle and the preceding vehicle and the following vehicle respectively exceed the safety distance threshold, the vehicle is controlled to honk.

6. A vehicle control device based on side parking of the preceding vehicle, characterized in that: The method comprises a unit for executing the method according to any one of claims 1 to 5.

7. A vehicle, characterized in that: The vehicle includes a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the method according to any one of claims 1 to 5 is implemented.

8. A computer-readable storage medium, characterized in that The storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 5 can be implemented.

Images