Travel control method, travel control device, and intelligent driving vehicle

Abstract

The present disclosure provides a driving control method, a driving control device and an intelligent driving vehicle, and relates to the field of computers, in particular to the field of intelligent driving. The driving control method comprises: in response to an emergency braking decision, determining trajectory points of an intelligent driving vehicle and a target triggering the emergency braking decision within a specified future time period; determining whether the target and the intelligent driving vehicle have a collision risk according to the trajectory points of the intelligent driving vehicle and the target triggering the emergency braking decision within the specified future time period; in the case of determining that there is a collision risk, maintaining the emergency braking decision, and in the case of determining that there is no collision risk, canceling the emergency braking decision. After the emergency braking decision, an additional trajectory-based emergency braking decision filtering method is introduced to maintain or cancel the emergency braking decision, avoid the decision conflict between the AEB system and the PNC system, reduce the false triggering behavior of the emergency braking of the AEB system, and improve the operation efficiency of the intelligent driving vehicle.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the field of computers, in particular to the field of intelligent driving, and specifically to a driving control method, a driving control device and an intelligent driving vehicle. BACKGROUND

[0002] An intelligent driving vehicle is used to automatically transport people or objects from one location to another location by collecting environmental information through sensors on the vehicle and completing automatic transportation. Intelligent driving vehicles controlled based on autonomous driving technology perform logistics transportation, greatly improving the convenience of production and life and saving labor costs.

[0003] An active safety module is an important component in an autonomous driving system. An autonomous emergency braking (AEB) system of the active safety module can perceive and decide the driving environment, and issue an emergency braking instruction to the vehicle chassis when a dangerous situation occurs to ensure the safety of the vehicle. The autonomous driving system also includes a planning and control (PNC) system responsible for planning the driving path of the vehicle and controlling the actions of the vehicle.

[0004] The AEB system and the PNC system may have a decision conflict problem. For example, the AEB system judges based on the environmental information at the current time that there is a conflict risk between the vehicle and the target, and needs to perform emergency braking. However, the PNC system considers that there is no conflict risk and that obstacle avoidance behavior can maintain normal driving. In this case, the emergency braking performed by the AEB system is a false trigger, which originally does not need to trigger emergency braking, greatly affecting the operation efficiency of the intelligent driving vehicle. SUMMARY

[0005] To solve the decision conflict problem between the AEB system and the PNC system, the responses made by the AEB system and the PNC system based on the information obtained by each of them are correct and cannot be further optimized. Embodiments of the present disclosure introduce an additional trajectory-based emergency braking decision filtering method after the emergency braking decision to maintain or cancel the emergency braking decision, avoid the decision conflict between the AEB system and the PNC system, reduce the false trigger behavior of the emergency braking of the AEB system, and improve the operation efficiency of the intelligent driving vehicle.

[0006] Some embodiments of the present disclosure provide a driving control method, including: in response to an emergency braking decision, determining trajectory points of an intelligent driving vehicle and a target triggering the emergency braking decision within a specified time period in the future; determining whether there is a conflict risk between the target and the intelligent driving vehicle according to the trajectory points of the intelligent driving vehicle and the target triggering the emergency braking decision within the specified time period in the future; in the case of determining that there is a conflict risk, maintaining the emergency braking decision, and in the case of determining that there is no conflict risk, canceling the emergency braking decision.

[0007] In some embodiments, determining whether the target has a collision risk with the intelligent driving vehicle according to the trajectory points of the target triggering the emergency braking decision and the intelligent driving vehicle within a specified time period in the future comprises: if the trajectory point of the target triggering the emergency braking decision at any time within the specified time period in the future is located within the vehicle frame range of the trajectory point of the intelligent driving vehicle at the time, determining that the target has a collision risk with the intelligent driving vehicle.

[0008] In some embodiments, determining whether the target has a collision risk with the intelligent driving vehicle according to the trajectory points of the target triggering the emergency braking decision and the intelligent driving vehicle within a specified time period in the future comprises: if the line connecting the trajectory points of the target triggering the emergency braking decision within the specified time period in the future intersects the line connecting the trajectory points of the intelligent driving vehicle within the specified time period in the future, determining that the target has a collision risk with the intelligent driving vehicle.

[0009] In some embodiments, determining whether the target has a collision risk with the intelligent driving vehicle according to the trajectory points of the target triggering the emergency braking decision and the intelligent driving vehicle within a specified time period in the future comprises: determining whether the trajectory point of the target triggering the emergency braking decision at any time within the specified time period in the future is located within the vehicle frame range of the trajectory point of the intelligent driving vehicle at the time; in the case of a positive determination, determining that the target has a collision risk with the intelligent driving vehicle; in the case of a negative determination, determining whether the line connecting the trajectory points of the target triggering the emergency braking decision within the specified time period in the future intersects the line connecting the trajectory points of the intelligent driving vehicle within the specified time period in the future, and if the determination is that the lines intersect, determining that the target has a collision risk with the intelligent driving vehicle.

[0010] In some embodiments, determining whether the target has a collision risk with the intelligent driving vehicle according to the trajectory points of the target triggering the emergency braking decision and the intelligent driving vehicle within a specified time period in the future comprises: in the case of determining no collision risk according to the trajectory points, again determining whether to perform emergency braking according to the motion state information of the last trajectory point of the target triggering the emergency braking decision and the intelligent driving vehicle within the specified time period in the future; if the result of the re-determination is still that emergency braking is needed, determining that the target has a collision risk with the intelligent driving vehicle, and if the result of the re-determination is that emergency braking is not needed, determining that the target has no collision risk with the intelligent driving vehicle.

[0011] In some embodiments, the re-determination of whether to perform emergency braking comprises: determining whether the target has a collision risk with the intelligent driving vehicle according to the position, speed and acceleration of the target in the intelligent driving vehicle coordinate system at the time of the last trajectory point, and the speed and braking properties of the intelligent driving vehicle; and in the case of determining a collision risk, determining that emergency braking needs to be performed.

[0012] In some embodiments, determining whether the target has a collision risk with the intelligent driving vehicle comprises: determining a relative speed of the target relative to the intelligent driving vehicle at the last trajectory point, and determining whether the target is a candidate target triggering emergency braking according to whether a ray of the target in a direction of the relative speed passes through a safe driving range of the intelligent driving vehicle; in a case where the target is the candidate target triggering emergency braking, determining whether the target has a collision risk with the intelligent driving vehicle according to a position, a speed and an acceleration of the target in a coordinate system of the intelligent driving vehicle at the last trajectory point, and a speed and braking attributes of the intelligent driving vehicle.

[0013] In some embodiments, determining whether the target has a collision risk with the intelligent driving vehicle comprises: determining a braking distance of the target and a braking distance of the intelligent driving vehicle according to a radial component of the speed and a radial component of the acceleration of the target at the last trajectory point, and the speed and the braking attributes of the intelligent driving vehicle; and determining whether the target has a collision risk with the intelligent driving vehicle in a radial direction by comparing a difference between the braking distance of the intelligent driving vehicle and the braking distance of the target with a radial distance of the target from the intelligent driving vehicle, wherein the radial distance of the target from the intelligent driving vehicle is determined according to a radial component of a position of the target in the coordinate system of the intelligent driving vehicle.

[0014] In some embodiments, determining whether the target has a collision risk with the intelligent driving vehicle comprises: determining whether the target will move into a width range of the intelligent driving vehicle in a lateral direction according to a lateral component of the position, the speed and the acceleration of the target in the coordinate system of the intelligent driving vehicle at the last trajectory point, and the speed of the intelligent driving vehicle; and determining whether the target has a collision risk with the intelligent driving vehicle in the lateral direction according to whether the target will move into the width range of the intelligent driving vehicle in the lateral direction.

[0015] Some embodiments of the present disclosure provide a driving control device, comprising: a memory; and a processor coupled to the memory, the processor being configured to execute a driving control method based on instructions stored in the memory.

[0016] Some embodiments of the present disclosure provide an intelligent driving vehicle, comprising: a driving control device configured to execute a driving control method.

[0017] Some embodiments of the present disclosure provide a computer-readable storage medium having stored thereon computer instructions, which, when executed by a processor, implement steps of a driving control method.

[0018] Some embodiments of the present disclosure provide a computer program product comprising computer instructions, which, when executed by a processor, implement steps of a driving control method. BRIEF DESCRIPTION OF DRAWINGS

[0019] The accompanying drawings that are needed to be used in the embodiments or related technical descriptions will be briefly introduced below. The present disclosure can be more clearly understood according to the detailed description below with reference to the accompanying drawings.

[0020] It is obvious that the accompanying drawings in the following description are only some embodiments of the present disclosure, and other accompanying drawings can be obtained by those skilled in the art without creative labor on the basis of the accompanying drawings.

[0021] Figure 1 An electrical architecture schematic diagram of an intelligent (automatic) driving vehicle is shown.

[0022] Figure 2 An external structure schematic diagram of an intelligent (automatic) driving vehicle is shown.

[0023] Figure 3 A schematic diagram of a neural network for processing point clouds is shown.

[0024] Figure 4 A schematic diagram of AEB system and PNC system decision conflict is shown.

[0025] Figure 5 A flowchart of a driving control method is shown.

[0026] Figure 6 A schematic diagram of a trajectory point-based conflict risk determination method is shown.

[0027] Figure 7 A schematic diagram of a trajectory line-based conflict risk determination method is shown.

[0028] Figure 8 A schematic diagram of a conflict risk determination situation based on emergency braking re-judgment is shown.

[0029] Figure 9 A flowchart of a driving control method is shown.

[0030] Figure 10 A flowchart of an emergency braking decision method is shown.

[0031] Figure 11 A schematic diagram of determining whether a target is a candidate target for triggering emergency braking is shown.

[0032] Figure 12 、 Figure 13 and Figure 14A schematic diagram showing a target braking distance and a braking distance of an intelligent driving vehicle in different situations.

[0033] Figure 15 A structural schematic diagram of a travel control device showing some embodiments of the present disclosure. DETAILED DESCRIPTION

[0034] It should be noted that the relative arrangement, numerical expressions, and numerical values of the components and steps set forth in these embodiments do not limit the scope of the present disclosure unless specifically stated otherwise.

[0035] Those skilled in the art can understand that the terms "first", "second", etc. in the embodiments of the present disclosure are only used to distinguish different steps, devices or modules, and do not represent any specific technical meaning, nor do they represent the inevitable logical order between them.

[0036] It should also be understood that in the embodiments of the present disclosure, "a plurality of" can mean two or more, and "at least one" can mean one, two or more.

[0037] It should also be understood that for any component, data or structure mentioned in the embodiments of the present disclosure, unless specifically limited or given a contrary implication by the context, it can be understood as one or more in general.

[0038] In addition, the term "and / or" in the present disclosure is only a description of the association relationship between the associated objects, which means that there can be three relationships, for example, A and / or B can mean that there are three cases of A alone, A and B together, and B alone. In addition, the character " / " in the present disclosure generally represents an "or" relationship between the front and rear associated objects.

[0039] It should also be understood that the description of various embodiments of the present disclosure emphasizes the differences between various embodiments, and the same or similar parts can be referred to each other, and for the sake of brevity, will not be repeated.

[0040] At the same time, it should be understood that in order to facilitate description, the size of each part shown in the drawings is not drawn according to the actual proportional relationship.

[0041] The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way limiting on the disclosure, its application or uses.

[0042] Techniques, methods, and devices known to those of ordinary skill in the relevant art can not be discussed in detail, but should be considered part of the specification where appropriate.

[0043] It should be noted that like reference numerals and characters refer to like elements throughout the several views of the drawings, and that, unless otherwise indicated, like reference numerals and characters refer to like elements throughout the several views of the drawings.

[0044] Moreover, for the avoidance of obscuring the present disclosure with unnecessary detail, only the processing steps and / or device structures that are most closely related to the schemes according to the present disclosure are shown in the drawings, while other details that are less relevant to the present disclosure are omitted. It should also be noted that like reference numerals and characters refer to like elements throughout the several views of the drawings, and that, once an element is defined in one drawing, it need not be discussed further in subsequent drawings.

[0045] Figure 1 An electrical architecture schematic diagram of an intelligent (autonomous) driving vehicle showing some embodiments of the present disclosure.

[0046] As Figure 1 shown, the intelligent (autonomous) driving vehicle 100 of this embodiment, for example, includes an autonomous driving module 110 and a chassis module 120, and, as needed, can also include a remote monitoring push module 130 and a cargo box module 140. For example, a vehicle that has a remote monitoring requirement is provided with the remote monitoring push module 130, and a vehicle that does not have a remote monitoring requirement can not be provided with the remote monitoring push module 130. For another example, a vehicle that has a cargo carrying requirement (such as a truck) is provided with the cargo box module 140, and a vehicle that does not have a cargo carrying requirement (such as a people-carrying car) can not be provided with the cargo box module 140. The intelligent (autonomous) driving vehicle can be, for example, a driverless vehicle, a driverless delivery vehicle, a driverless vending vehicle, etc.

[0047] The automatic driving module 110 includes one or more of a central processor (Orin or Xavier module) 111, a traffic light recognition camera 112, a front side camera 1131, a rear side camera 1132, a left side camera 1133, a right side camera 1134, a laser radar 114, a front side blind filling radar 1151, a rear side blind filling radar 1152, a left side blind filling radar 1153, a right side blind filling radar 1154, a positioning module (such as Beidou, GPS, etc.) 116, an inertial navigation unit 117, a switch 118, etc. as needed. The cameras can communicate with the automatic driving module. In order to improve the transmission speed and reduce the wire harness, GMSL (Gigabit Multimedia Serial Links) link communication can be used. The central processor 111 can be implemented by a general central processor, a digital signal central processor (Digital Signal Processor, DSP), an application specific integrated circuit (Application Specific Integrated Circuit, ASIC), a field programmable gate array (Field Programmable Gate Array, FPGA) or other programmable logic device, a discrete gate or transistor, etc. The central processor 111 can be configured to perform automatic driving control.

[0048] The chassis module 120 includes one or more of a battery 121, a power management device 122, a chassis controller 123, a motor driver 124, a power motor 125, a communication module 126 as needed. The battery 121 provides power for the entire automatic driving vehicle system. The battery 121 includes a main battery 1211 and a standby battery 1212. When the automatic driving vehicle is running, the main battery 1211 supplies power to each module of the automatic driving vehicle. When the automatic driving vehicle is on standby, the standby battery 1212 supplies power to the central processor 111 and the communication module 126. The power management device 122 converts the output of the battery 121 into different voltage levels that can be used by each module, and controls power-on and power-off. The chassis controller 123 receives the motion instructions issued by the automatic driving module 110, and controls the steering, forward movement, backward movement, braking, etc. of the automatic driving vehicle. The communication module 126 communicates with the background server, and can realize remote control of the automatic driving vehicle by the background operator. The communication module 126 includes a cellular wireless communication device 1261 and a radio frequency communication device 1262. The cellular wireless communication device 1261 communicates using cellular wireless communication technology, such as 2G (second generation), 3G (third generation), 4G (fourth generation) or 5G (fifth generation) cellular wireless communication technology. The radio frequency communication device 1262 communicates using radio frequency communication technology.

[0049] The remote monitoring streaming module 130 may include, as needed, one or more of the following: a front monitoring camera 1311, a rear monitoring camera 1312, a left monitoring camera 1313, a right monitoring camera 1314, and a streaming module 132. The streaming module 132 transmits the video data captured by the monitoring cameras 1311-1314 to the backend server for viewing by backend operators.

[0050] The cargo box module 140 may include a cargo box 141 as needed, which is a cargo-carrying device for the autonomous vehicle. The cargo box module 140 also includes a display and interaction module 142 for interaction between the autonomous vehicle and the user. Users can perform operations such as picking up items, storing goods, and purchasing goods through the display and interaction module 142. The type of cargo box 141 can be changed according to actual needs. For example, in a logistics scenario, the cargo box may include multiple sub-boxes of different sizes, which can be used to load goods for delivery. In a retail scenario, the cargo box can be set as a transparent box so that users can see the products for sale.

[0051] Figure 2 The diagram shows the external structure of an intelligent (autonomous) driving vehicle according to some embodiments of the present disclosure. Figure 2 The diagram shows the external structure of an autonomous vehicle capable of carrying cargo. Autonomous vehicles with different functions can have different external structures; for example, the external structure of a passenger-carrying autonomous vehicle can be referenced from that of a car. Figure 2 As shown, from the current perspective, the chassis 210, cargo box 141, display and interaction module 142, right-side camera 1134, lidar 114, rear blind spot radar 1152, left-side blind spot radar 1153, and right-side blind spot radar 1154 of the autonomous vehicle 200 can be seen.

[0052] The active safety module of the intelligent driving vehicle in this disclosure embodiment uses lidar signals as input to perceive surrounding targets and their motion states. LiDAR is denser than millimeter-wave radar, can effectively perceive small targets, and has higher vertical resolution, enabling the separation of static obstacles from the background. Based on the solution proposed in this disclosure, the following will combine... Figure 3 The description shows that by using time-series lidar signals, the speed, acceleration, and category information of a target can be perceived. Furthermore, the speed signal predicted by the two-dimensional speed prediction map has high accuracy in both the radial and lateral directions, which can reflect the true motion state of the target and provide rich and high-quality upstream information for the active safety decision-making of autonomous driving systems.

[0053] Figure 3 Schematic diagrams of neural networks for processing point clouds, representing some embodiments of this disclosure, are shown. These point cloud processing neural networks can exist in the product shape of a point cloud processing device. For example... Figure 3As shown, the point cloud processing neural network (point cloud processing apparatus) comprises: a feature extraction module 310 configured to perform feature extraction on the point cloud data of the current frame of the lidar to obtain a point cloud feature map of the current frame; a temporal fusion module 320 configured to perform temporal fusion processing on the point cloud feature map of the current frame and the fusion feature map of the previous frame, i.e., to splice the point cloud feature map of the current frame and the fusion feature map of the previous frame and extract temporal features to obtain a fusion feature map of the current frame; and a prediction module 330 configured to predict one or more of the speed, acceleration and category of the target according to the fusion feature map of the current frame to obtain a speed prediction map, an acceleration prediction map and a category prediction map.

[0054] The feature extraction module 310, for example, comprises a downsampling network, a plurality of residual networks, a plurality of upsampling networks with different step sizes, and a splicing unit. The downsampling network performs downsampling processing on the point cloud data of the current frame of the lidar to output a first feature map of the current frame. The plurality of cascaded residual networks perform downsampling processing on the first feature map to output a plurality of second feature maps with different downsampling rates. The plurality of upsampling networks with different step sizes perform upsampling processing on the plurality of second feature maps with different downsampling rates respectively to output a plurality of third feature maps with the same size. The splicing unit can splice the plurality of third feature maps to output the point cloud feature map of the current frame.

[0055] An example of the feature extraction module 310 is listed below. As shown in FIG. 3, the feature extraction module 310 comprises a downsampling network 3101, a plurality of residual networks 3102, a plurality of upsampling networks with different step sizes 3103, and a splicing unit 3104. Figure 3As shown, the point cloud information of the current frame (T time) is voxelized to form voxelized point cloud data as input data. The input data is denoted as [H, W, D], where H, W, and D represent height, width, and depth, respectively. The input data is first reduced in feature size by a downsampling network with a step size of 2 to reduce the computational load of the subsequent network. The feature map output by the downsampling network is denoted as [H / 2, W / 2, C0], the height and width are reduced to 1 / 2 of the original, and C0 represents the channel. The structure of the downsampling network is, for example, a convolutional layer with a step size of N (N = 2 in this example) (the convolutional layer performs downsampling) and with a BatchNorm (Batch Normalization) layer and a ReLU (Rectified Linear Unit) activation function layer. Then, the data output by the downsampling network will pass through four consecutive residual networks to extract deeper features. The convolutional step size of the four residual networks is, for example, 2, and they output feature maps with downsampling rates of 4 / 8 / 16 / 32, denoted as [H / 4, W / 4, C1], [H / 8, W / 8, C2], [H / 16, W / 16, C3], and [H / 32, W / 32, C4], with heights and widths reduced to 1 / 4, 1 / 8, 1 / 16, and 1 / 32 of the original, and C1, C2, C3, and C4 representing the channels, respectively. The residual network introduces cross-layer connections to add the input signal directly to the output of the residual network, making it easier to pass gradients during backpropagation, thereby solving the gradient vanishing problem in deep network training. Subsequently, the four feature maps of the four residual networks pass through up-sampling networks with step sizes of 1 / 2 / 4 / 8, respectively, to output four feature maps of the same size with a downsampling rate of 4, all denoted as [H / 4, W / 4, C / 4], with heights and widths of 1 / 4 of the original, and H, W, and C representing height, width, and channel, respectively. The structure of the up-sampling network is, for example, a deconvolutional layer with a step size of N (N = 2 in this example) (the deconvolutional layer performs up-sampling) and with a BatchNorm layer and a ReLU activation function layer. The four up-sampled feature maps of the same size represent feature information at different network depths. Shallow features mainly focus on local details and low-level features of the image, which are usually closely related to the pixels of the image, including color, texture, edge, and corner information. Deep features focus on global and high-level features, which are more abstract and complex representations of the features, which are usually related to the semantic information of the image. Finally, the multiple feature maps obtained after up-sampling are concatenated in the channel dimension by a concatenation unit to obtain the point cloud feature map of the current frame, denoted as [H / 4, W / 4, C].

[0056] The time sequence fusion module 320 includes, for example, a coordinate system alignment unit, a splicing unit, and a time sequence fusion neural network, and can also include an up-sampling network. The coordinate system alignment unit aligns the fusion feature map of the previous frame to the coordinate system of the current frame according to the coordinate transformation matrix between the previous frame and the current frame. The splicing unit splices the aligned fusion feature map of the previous frame and the point cloud feature map of the current frame in the channel dimension. The time sequence fusion neural network extracts time sequence features from the spliced feature map to obtain the fusion feature map of the current frame. The fusion feature map of the current frame includes time sequence information of each historical frame before the current frame, and the fusion feature map of the previous frame includes time sequence information of each historical frame before the previous frame. The fusion feature map of the initial frame is obtained by extracting time sequence features from the point cloud feature map of the initial frame by using the time sequence fusion neural network. On this basis, through iteration, the fusion feature map of each frame can be obtained. By introducing time sequence features, the target motion state can be better perceived. The up-sampling network performs up-sampling processing on the fusion feature map of the current frame to obtain a fusion feature map with the same size as the point cloud data of the current frame.

[0057] An example of the time sequence fusion module 320 is listed below. As shown in Figure 3 The fusion feature map (with a size of [H / 4, W / 4, C]) of the previous frame (at time T-1) point cloud is twisted and aligned to the coordinate system of the current frame (at time T) according to the coordinate transformation matrix between the two frames, and then is spliced with the point cloud feature map (with a size of [H / 4, W / 4, C]) of the current frame output by the feature extraction module 310, and is input into the time sequence fusion neural network. The network compares the two frame point cloud features, mines time sequence features, and outputs the fusion feature map (with a size of [H / 4, W / 4, C]) of the current frame. The time sequence fusion neural network includes a plurality of network blocks (such as 3) in cascade, each network block includes a convolution layer (with a step of 1), a batch normalization layer, and an activation layer such as a ReLU activation function layer. After the fusion feature map (with a size of [H / 4, W / 4, C]) of the current frame is up-sampled by four (with a step of 4), a fusion feature matrix (with a size of [H, W, C]) with the same size as the original image is obtained.

[0058] The prediction module 330 includes, for example, a speed prediction neural network, an acceleration prediction neural network, and a category prediction neural network. The speed prediction neural network, the acceleration prediction neural network, and the category prediction neural network process the fused feature map of the current frame respectively, and output a speed prediction map, an acceleration prediction map, and a category prediction map respectively, which have the same size as the original map (original point cloud voxel grid). The speed prediction map [H, W, 2] has a dimension of 2, representing the radial component and the transverse component of the speed of the target. In the vehicle driving plane, the radial direction refers to the driving direction of the vehicle, i.e., the head direction, and the transverse direction refers to the direction perpendicular to the driving direction of the vehicle, i.e., the width direction of the vehicle body. The acceleration prediction map [H, W, 1] has a dimension of 1, representing the rate of change of the speed of the target. The category prediction map [H, W, 1] has a dimension of 1, representing a one-dimensional index of the category of the target. The speed / acceleration / category prediction neural network includes multiple network blocks (such as 2) cascaded, and each network block includes a convolution layer (with a step of 1), a batch normalization layer, and an activation layer such as a ReLU activation function layer. The last network block is cascaded with a convolution layer (with a step of 1).

[0059] The point cloud processing neural network is usually trained and then used for prediction. During training, the speed prediction map and the acceleration prediction map can use an L1 loss function and be supervised by true values, and the category prediction map can use a Focal Loss loss function for supervised learning. The coefficients of the three loss functions can be set, for example, to 1.0, 2.0, and 4.0 respectively. Under this coefficient configuration, the speed, acceleration, and category information can be learned simultaneously, which is beneficial to the convergence of the point cloud processing neural network. Based on the point cloud processing neural network, the laser radar point cloud can be processed to obtain the speed, acceleration, and category of the target around the intelligent driving vehicle.

[0060] In addition, the laser radar can calculate the position coordinates of the target by measuring the time difference from the emission to the reception of the laser pulse, as well as the emission angle and the pitch angle of the laser beam. The position coordinates can be used to describe the specific position of the target in space.

[0061] Based on the upstream perception module, the laser radar and the point cloud processing neural network (point cloud processing device), the position, speed, acceleration, and category of the target around the intelligent driving vehicle can be perceived simultaneously.

[0062] As mentioned earlier, in the automatic driving system, the AEB system and the PNC system may have decision conflict problems. For example, Figure 4As shown, the obstacle appears in front of the intelligent driving vehicle, and the intelligent driving vehicle travels at a large speed. If the AEB system is determined according to the current instantaneous motion state, the AEB system performs emergency braking is correct, because the speed of the intelligent driving vehicle is large, the distance to the obstacle is short, and the intelligent driving vehicle cannot brake in time and needs emergency braking. However, in the view of the PNC system, the PNC system plans the intelligent driving vehicle to turn left around the obstacle at a future time, and thus the PNC system determines that the intelligent driving vehicle does not need to brake. In the above scenario, the AEB system and the PNC system make responses based on the information obtained by each system, and the responses are correct. However, this causes the AEB system to have a false triggering behavior of emergency braking in a scenario where the intelligent driving vehicle can turn around but cannot brake in time.

[0063] The embodiment of the present disclosure additionally introduces a trajectory-based emergency braking decision filtering method (referred to as a "trajectory filtering decision algorithm") after the emergency braking decision, which supplements the emergency braking decision algorithm of the AEB system. If the emergency braking decision algorithm determines that emergency braking is not needed, the trajectory filtering decision algorithm does not need to be started. If the emergency braking decision algorithm determines that emergency braking is needed, the trajectory filtering decision algorithm is started again to determine whether to maintain or cancel the emergency braking decision. Only the determination of the trajectory filtering decision algorithm can maintain the emergency braking decision triggered by the AEB system.

[0064] The input information of the "trajectory filtering decision algorithm" of the embodiment includes: (1) the motion state information of the target triggering the emergency braking in the upstream AEB system, for example, the position (x, y), speed (vx, vy), acceleration (ax, ay), and category (optional, set as objType) of the target in the intelligent driving vehicle coordinate system. According to the motion state information of the target, the trajectory points of the target in a specified future time period can be calculated. The intelligent driving vehicle coordinate system is, for example, with the center of the vehicle as the origin, with the head direction as the radial direction (x direction), and with the width direction of the vehicle body as the transverse direction (y direction). It is assumed that: the head direction is the radial positive direction, the tail direction is the radial negative direction, and the left side is the transverse positive direction and the right side is the transverse negative direction. (2) The trajectory points of the intelligent driving vehicle (also referred to as the ego vehicle) in a specified future time period calculated by the chassis control module. The motion state information of the intelligent driving vehicle, such as the speed (set as ego_speed), is directly and real-timely obtained from the vehicle chassis, and does not need to be predicted by the perception module. The chassis control module can calculate the trajectory points of the intelligent driving vehicle in a specified future time period according to the motion state information of the intelligent driving vehicle. For example, the intelligent driving vehicle has 8 future trajectory points every 0.15s in the future 1.2s. Each trajectory point contains the position (ego_x, ego_y) and speed information (ego_speed, which can be decomposed into ego_vx, ego_vy) of the ego vehicle at this time.

[0065] Based on the various input information obtained above for the "trajectory filtering decision algorithm", the "trajectory filtering decision algorithm" can be executed to perform driving control.

[0066] Figure 5 A flowchart illustrating some embodiments of the driving control method of this disclosure is shown. Figure 5 As shown, the driving control method of this embodiment includes the following steps. This driving control method can, for example, be executed by a driving control device of an intelligent driving vehicle (hereinafter referred to as "the vehicle").

[0067] In step 510, in response to the upstream emergency braking decision, the trajectory points of the intelligent driving vehicle and the target that triggered the emergency braking decision are determined within a specified time period in the future.

[0068] As mentioned earlier, based on the target's motion state information, the target's trajectory points within a specified future time period can be calculated; similarly, based on the autonomous vehicle's motion state information, the autonomous vehicle's trajectory points within a specified future time period can be calculated. Thus, the trajectory points of both the autonomous vehicle and the target that triggered the emergency braking decision can be obtained within the specified future time period.

[0069] In step 520, based on the trajectory points of the intelligent driving vehicle and the target that triggers the emergency braking decision within a specified time period in the future, it is determined whether there is a risk of conflict between the target and the intelligent driving vehicle.

[0070] In some embodiments, step 520 may involve performing one or more of steps 520a, 520b, and 520c to determine whether there is a risk of conflict between the target and the autonomous vehicle. For example, step 520a may be performed first, followed by step 520b, and then step 520c.

[0071] Step 520a, a method for determining conflict risk based on trajectory points. (For example...) Figure 6 As shown, if the trajectory point of the target that triggers the emergency braking decision is located within the vehicle frame of the intelligent driving vehicle's trajectory point at any time within a specified future time period, it is determined that there is a risk of conflict between the target and the intelligent driving vehicle.

[0072] Step 520b, a method for determining conflict risk based on trajectory lines. (For example...) Figure 7 As shown, if the line connecting the trajectory points of the target that triggers the emergency braking decision within a specified future time period intersects with the line connecting the trajectory points of the autonomous vehicle within the same specified future time period, a conflict risk between the target and the autonomous vehicle is determined. This identifies the conflict risk arising from the target "crossing" the autonomous vehicle.

[0073] Step 520c, a conflict risk determination method based on emergency braking reassessment. For example... Figure 8As shown, the judgment of emergency braking is made again according to the intelligent driving vehicle and the motion state information of the last trajectory point of the target triggering the emergency braking decision within a specified time period in the future; if the re-judgment result is still that emergency braking is needed, it is determined that the target has a collision risk with the intelligent driving vehicle, and if the re-judgment result is that emergency braking is not needed, it is determined that the target has no collision risk with the intelligent driving vehicle. Thus, the collision risk close to the future specified time period is identified. In addition, the following will combine Figure 10 An emergency braking decision method is described.

[0074] In step 530, in the case of determining a collision risk, the emergency braking decision is maintained, and in the case of determining no collision risk, the emergency braking decision is cancelled.

[0075] In the embodiments of the present disclosure, after the emergency braking decision, an additional trajectory-based emergency braking decision filtering method is introduced to maintain or cancel the emergency braking decision, avoid the decision conflict between the AEB system and the PNC system, reduce the false triggering behavior of the emergency braking of the AEB system, and improve the operation efficiency of the intelligent driving vehicle.

[0076] For example, in the scenario of “can detour but not in time to brake”, before the introduction of the “trajectory filtering decision algorithm”, the emergency braking will be falsely triggered, and after the introduction of the “trajectory filtering decision algorithm”, the emergency braking decision of the AEB system will be cancelled, avoiding the decision conflict between the AEB system and the PNC system.

[0077] The following will combine Figure 9 An example of a driving control method is described. The steps in this example and the execution order thereof are only one example of driving control, and the steps and the execution order thereof can also be executed in other methods.

[0078] Figure 9 A flowchart of a driving control method of some embodiments of the present disclosure is shown. As shown, the driving control method of this embodiment includes the following steps. The driving control method may, for example, be executed by a driving control device of an intelligent driving vehicle (referred to as “vehicle”). Figure 9

[0079] In step 910, in response to the upstream emergency braking decision, the trajectory points of the intelligent driving vehicle and the target triggering the emergency braking decision within a specified time period in the future are determined.

[0080] In step 920, it is judged whether the trajectory point of the target triggering the emergency braking decision within a specified time period in the future is located within the vehicle frame range of the trajectory point of the intelligent driving vehicle at that time. If yes, it is determined that the target has a collision risk with the intelligent driving vehicle, and step 950a is executed, and if no, step 930 is executed to continue the judgment.

[0081] ​At step 930, it is determined whether the line connecting each trajectory point of the target triggering the emergency braking decision in the future specified time period intersects with the line connecting each trajectory point of the intelligent driving vehicle in the future specified time period. If yes, it is determined that the target and the intelligent driving vehicle have a collision risk, and step 950a is executed; if not, step 940 is executed to continue determining.

[0082] At step 940, it is determined whether the emergency braking re-determination result is still to brake urgently. If yes, it is determined that the target and the intelligent driving vehicle have a collision risk, and step 950a is executed; if not, i.e., the re-determination result is to not brake urgently, it is determined that the target and the intelligent driving vehicle have no collision risk, and step 950b is executed.

[0083] According to the motion state information of the last trajectory point of the intelligent driving vehicle and the target triggering the emergency braking decision in the future specified time period, the emergency braking is re-determined.

[0084] At step 950a, in the case of determining that there is a collision risk, the emergency braking decision is maintained.

[0085] At step 950b, in the case of determining that there is no collision risk, the emergency braking decision is cancelled.

[0086] In the embodiments of the present disclosure, after the emergency braking decision, an additional trajectory-based emergency braking decision filtering method is introduced to maintain or cancel the emergency braking decision, avoid the decision conflict between the AEB system and the PNC system, reduce the false triggering behavior of the emergency braking of the AEB system, and improve the operation efficiency of the intelligent driving vehicle.

[0087] The method of re-determining the emergency braking will be described below in combination with Figure 10 . Figure 10 The flowchart of the emergency braking decision method of some embodiments of the present disclosure is shown. As shown in Figure 10 , the emergency braking decision method of this embodiment includes the following steps.

[0088] At step 1010, the position, speed and acceleration of the target in the intelligent driving vehicle coordinate system at the last trajectory point, and the speed and braking properties of the intelligent driving vehicle are obtained.

[0089] At step 1020, according to the position, speed and acceleration of the target in the intelligent driving vehicle coordinate system at the last trajectory point, and the speed and braking properties of the intelligent driving vehicle, it is determined whether the target and the intelligent driving vehicle have a collision risk.

[0090] Thus, based on the rich information such as the speed and acceleration of the perceived target, it is more accurately determined whether the target and the intelligent driving vehicle have a collision risk.

[0091] In some embodiments, step 1020 can be executed by performing step 1020c and step 1020d to determine whether the target has a collision risk with the intelligent driving vehicle, or at least one of step 1020a and step 1020b can be performed first to preliminarily exclude the collision risk, and then step 1020c and step 1020d can be performed to verify whether the target has a collision risk with the intelligent driving vehicle.

[0092] Step 1020a, preliminary judgment of collision risk. Determine whether the target has a potential collision risk with the intelligent driving vehicle based on at least one of the position of the target in the intelligent driving vehicle coordinate system, the speed of the target, and the speed of the intelligent driving vehicle. Thus, based on simple judgment logic, the potential collision risk is quickly determined, and the situation where there is no collision risk is excluded, and the situation where the collision risk cannot be excluded is further verified by subsequent judgment. In the case where it is determined that there is a potential collision risk, whether the target has a collision risk with the intelligent driving vehicle is determined based on the position, speed and acceleration of the target in the intelligent driving vehicle coordinate system, and the speed and braking property of the intelligent driving vehicle.

[0093] Step 1020b, determine whether the target is a candidate target for triggering emergency braking. Determine the relative speed of the target relative to the intelligent driving vehicle, and determine whether the target is a candidate target for triggering emergency braking based on whether a ray of the target in the direction of the relative speed passes through the safe driving range of the intelligent driving vehicle, and the subsequent combination Figure 11 Description. If it is not a candidate target for triggering emergency braking, there is no need for further judgment; if it is a candidate target for triggering emergency braking, the collision risk can be further verified by subsequent judgment. In the case where the target is a candidate target for triggering emergency braking, whether the target has a collision risk with the intelligent driving vehicle is determined based on the position, speed and acceleration of the target in the intelligent driving vehicle coordinate system, and the speed and braking property of the intelligent driving vehicle.

[0094] Step 1020c, determine whether there is a collision risk in the radial direction, and the subsequent combination Figures 12-14 Description, in order to verify the radial collision risk situation.

[0095] Step 1020d, determine whether there is a collision risk in the lateral direction, in order to verify the lateral collision risk situation.

[0096] In step 1030, in the case where it is determined that there is a collision risk, it is determined that emergency braking needs to be performed. If it is determined that there is no collision risk, there is no need for emergency braking, and normal driving can continue.

[0097] Thus, based on the rich information such as the speed and acceleration of the target perceived, the collision risk of the target with the intelligent driving vehicle is more accurately determined, and then the decision whether to need emergency braking is accurately made.

[0098] The method of preliminarily judging the conflict risk of step 1020a is described below.

[0099] In some embodiments, according to the radial component of the position of the target in the intelligent driving vehicle coordinate system, it is determined that the target is located behind or in front of the intelligent driving vehicle; the radial component of the speed of the target is compared with the speed of the intelligent driving vehicle; in the case that the target is located behind the intelligent driving vehicle and the radial component of the speed of the target is less than the speed of the intelligent driving vehicle, it is determined that the target has no potential conflict risk with the intelligent driving vehicle; in the case that the target is located in front of the intelligent driving vehicle and the radial component of the speed of the target is greater than the speed of the intelligent driving vehicle, it is determined that the target has no potential conflict risk with the intelligent driving vehicle.

[0100] For example, if x < front_edge_to_center and vx < ego_speed, it means that the target is located behind the intelligent driving vehicle and the radial component of the speed of the target is less than the speed of the intelligent driving vehicle, at this time, the target is located at the rear side of the vehicle and gradually lags behind the vehicle and gradually increases the distance of lag, it can be determined that the target has no potential conflict risk with the intelligent driving vehicle, and emergency braking does not need to be triggered. If x > front_edge_to_center and vx > ego_speed, it means that the target is located in front of the intelligent driving vehicle and the radial component of the speed of the target is greater than the speed of the intelligent driving vehicle, at this time, the target will gradually move away from the host vehicle in the radial direction, so as to determine that the target has no potential conflict risk with the intelligent driving vehicle, and emergency braking does not need to be triggered.

[0101] In some embodiments, according to the lateral component of the position of the target in the intelligent driving vehicle coordinate system, it is determined that the target is located on the left or right of the intelligent driving vehicle; in the case that the target is located on the left of the intelligent driving vehicle and the lateral component of the speed of the target is to the left, it is determined that the target has no potential conflict risk with the intelligent driving vehicle; in the case that the target is located on the right of the intelligent driving vehicle and the lateral component of the speed of the target is to the right, it is determined that the target has no potential conflict risk with the intelligent driving vehicle.

[0102] For example, if y > right_edge_to_center and vy > 0, it means that the target is located on the left of the intelligent driving vehicle and the lateral component of the speed of the target is to the left, at this time, the target will gradually move away from the vehicle in the lateral direction to the left, it can be determined that the target has no potential conflict risk with the intelligent driving vehicle, and emergency braking does not need to be triggered. If y < -front_edge_to_center and vy < 0, it means that the target is located on the right of the intelligent driving vehicle and the lateral component of the speed of the target is to the right, at this time, the target will gradually move away from the vehicle in the lateral direction to the right, it can be determined that the target has no potential conflict risk with the intelligent driving vehicle, and emergency braking does not need to be triggered.

[0103] In some embodiments, according to the radial component of the position of the target in the intelligent driving vehicle coordinate system, it is determined whether the target is located behind the center point of the intelligent driving vehicle; in the case where the target is located behind the center point of the intelligent driving vehicle, it is determined that the target has no potential conflict risk with the intelligent driving vehicle.

[0104] For example, if x < 0, it means that the target is located behind the center point of the intelligent driving vehicle, and at this time the target is located behind the vehicle, it can be determined that the target has no potential conflict risk with the intelligent driving vehicle, and the emergency braking does not need to be triggered. The active safety module generally only processes targets that are level or in front of the vehicle, and does not need to process rear targets, otherwise it will cause more false triggering of emergency braking, affecting the vehicle passing efficiency and driving experience.

[0105] Therefore, according to the position of the target, the speed of the target and the speed of the intelligent driving vehicle can also be combined to quickly judge the potential conflict risk through simple judgment logic, exclude the case where there is no conflict risk, and the case where the conflict risk cannot be excluded, and then further verify the conflict risk through subsequent judgment.

[0106] The method of step 1020b for determining whether the target is a candidate target for triggering emergency braking is described below.

[0107] As shown in Figure 11 , the relative speed of the target relative to the intelligent driving vehicle is determined, and a ray is drawn from the target position with the relative speed direction as the direction of approach. If the ray of the target in the relative speed direction passes through the safe driving range of the intelligent driving vehicle, it means that the target is approaching the vehicle, and there is no conflict risk, and the target is determined to be a candidate target for triggering emergency braking. If the ray of the target in the relative speed direction does not pass through the safe driving range of the intelligent driving vehicle, it is determined that the target is not a candidate target for triggering emergency braking. The safe driving range of the intelligent driving vehicle can be a boundary box of the intelligent driving vehicle with a certain safety buffer distance, which is represented by an extended box. If it is not a candidate target for triggering emergency braking, it does not need to be further judged; if it is a candidate target for triggering emergency braking, the conflict risk can be further verified through subsequent judgment.

[0108] The method of step 1020c for determining whether there is a conflict risk in the radial direction is described below.

[0109] In some embodiments, the braking distances of the target and the autonomous vehicle are determined based on the radial components of the target's velocity and acceleration, as well as the speed and braking attributes of the autonomous vehicle. By comparing the difference between the braking distance of the autonomous vehicle and the target with the radial distance between the target and the autonomous vehicle, it is determined whether there is a risk of conflict between the target and the autonomous vehicle in the radial direction, in order to verify the radial conflict risk situation. If the difference between the braking distance of the autonomous vehicle and the target is greater than the difference between the radial distance between the target and the autonomous vehicle and the safety buffer distance, it is determined that there is a risk of conflict between the target and the autonomous vehicle in the radial direction. The radial distance between the target and the autonomous vehicle is determined based on the radial component of the target's position in the autonomous vehicle's coordinate system.

[0110] Due to system transmission delay, the autonomous vehicle first travels at the current speed for the duration of the system transmission delay (time_delay), and then performs emergency braking according to the slope of the maximum braking deceleration (ego_max_dec)d. The braking distance of the autonomous vehicle is the area enclosed by the solid line and the x-axis from t=0 to t=t_ego. If the target is traveling with a certain acceleration, the target's braking distance is the area enclosed by the dashed line and the x-axis from t=0 to t=t_target, with positive values ​​above the x-axis and negative values ​​below. The following sections determine the target's braking distance and the autonomous vehicle's braking distance in three different scenarios.

[0111] The first scenario: (e.g.) Figure 12 As shown, the target's initial velocity is in the same direction as the autonomous vehicle (vx > 0), and their velocities will not be the same at any given moment as the autonomous vehicle decelerates to 0. In this case, t_ego is the moment when the autonomous vehicle decelerates to 0, t_target is the moment when the target decelerates to 0, and the slope (acceleration) of the velocity v-time t graph is the target's acceleration.

[0112] If the radial components of the velocity of the target and the autonomous vehicle are in the same direction, and if the velocities of the autonomous vehicle and the target will not reach the same at any time during the deceleration process of the autonomous vehicle at its maximum braking deceleration, the distance traveled by the autonomous vehicle from its current speed to zero at its maximum braking deceleration is determined as the braking distance of the autonomous vehicle, and the distance traveled by the target from its current speed radial component to zero at its acceleration radial component is determined as the braking distance of the target.

[0113] The second scenario: Figure 13As shown, the initial speed of the target is in the same direction as the intelligent driving vehicle (vx > 0), and the speeds of the two will be the same at a certain time during the process of the intelligent driving vehicle reducing to 0. For this case, t_ego and t_target are both the time when the speeds of the two are the same, and the slope (acceleration) of the target in the speed-v-time graph is the acceleration of the target.

[0114] In the case where the radial component of the speed of the target is in the same direction as the intelligent driving vehicle, if the speeds of the intelligent driving vehicle and the target are the same at a certain time during the process of the intelligent driving vehicle reducing at the maximum braking deceleration, the driving distance of the intelligent driving vehicle during the process of reducing from the current speed at the maximum braking deceleration to the time when the speeds are the same is determined as the braking distance of the intelligent driving vehicle, and the driving distance of the target during the process of traveling from the current radial component of the speed at the acceleration radial component to the time when the speeds are the same is determined as the braking distance of the target.

[0115] The third case: as shown in Figure 14 the initial speed of the target is in the opposite direction of the intelligent driving vehicle (vx < 0). For this case, t_ego is the time when the intelligent driving vehicle reduces to 0, t_target is the time when the target reduces to 0, and the slope (acceleration) of the target in the speed-v-time graph can be a preset maximum deceleration target_max_dec set artificially in advance, rather than the deceleration in the real motion state of the target perceived. This is because the target opposite driving case is special, and it is necessary to assume that the other party immediately performs braking at the maximum deceleration. If there is danger under this assumption, the intelligent driving vehicle needs to take emergency braking.

[0116] In the case where the radial component of the speed of the target is in the opposite direction of the intelligent driving vehicle, the driving distance of the intelligent driving vehicle during the process of reducing from the current speed at the maximum braking deceleration to 0 is determined as the braking distance of the intelligent driving vehicle, and the driving distance of the target during the process of reducing from the current radial component of the speed at the preset maximum deceleration to 0 is determined as the braking distance of the target.

[0117] After determining the braking distance of the intelligent driving vehicle and the braking distance of the target, it is determined that the target and the intelligent driving vehicle have a risk of collision in the radial direction if the difference between the braking distance of the intelligent driving vehicle and the braking distance of the target is greater than the difference between the radial distance of the target from the intelligent driving vehicle and the safety buffer distance. The radial distance of the target from the intelligent driving vehicle is determined according to the position of the target in the intelligent driving vehicle coordinate system. That is, if “the braking distance of the intelligent driving vehicle - the braking distance of the target > the radial distance (x) of the target from the intelligent driving vehicle - the safety buffer distance”, it is determined that there is a risk of collision in the radial direction, and it is further determined whether there is a risk of collision in the lateral direction, otherwise, it is determined that there is no risk of collision in the radial direction, and the emergency braking does not need to be triggered.

[0118] The method for determining whether there is a risk of collision in the lateral direction in step 1020d is described below.

[0119] According to the position, speed and acceleration of the target in the intelligent driving vehicle coordinate system and the speed of the intelligent driving vehicle, it is determined whether the target will move into the width range of the intelligent driving vehicle in the lateral direction; according to whether the target will move into the width range of the intelligent driving vehicle in the lateral direction, it is determined whether the target and the intelligent driving vehicle have a risk of collision in the lateral direction, so as to verify the lateral collision risk. If the target will move into the width range of the intelligent driving vehicle in the lateral direction, it is determined that the target and the intelligent driving vehicle have a risk of collision in the lateral direction.

[0120] In some embodiments, the method of determining whether the target will move into the width range of the ego vehicle in the lateral direction comprises: determining the time of collision between the target and the ego vehicle in the radial direction according to the radial distance of the target from the ego vehicle, and the difference between the speed of the ego vehicle and the radial component vx of the speed of the target, i.e., calculating the time of collision in the radial direction t according to t = x / (ego_speed - vx), wherein x is the radial distance of the target from the ego vehicle determined according to the position of the target in the coordinate system of the ego vehicle; determining the position of the target in the lateral direction at the time of collision according to the current position of the target in the lateral direction y, the lateral component vy of the speed of the target, the lateral component ay of the acceleration of the target, and the time of collision t, i.e., determining the position of the target in the lateral direction y' at the time of collision according to y' = y + vy×t + ay×t×t / 2; and determining whether the target will move into the width range of the ego vehicle in the lateral direction according to the position of the target in the lateral direction at the time of collision, i.e., if y' > - right_edge_to_center and y' < left_edge_to_center, it means that the target will appear in the width range of the ego vehicle.

[0121] Figure 15 A structure diagram of a travel control device according to some embodiments of the present disclosure is shown. The ego vehicle comprises a travel control device, which can execute the travel control method in each embodiment.

[0122] As shown in Figure 15 the travel control device 1500 of this embodiment comprises a memory 1510 and a processor 1520 coupled to the memory 1510, and the processor 1520 is configured to execute the travel control method in any of the foregoing embodiments based on instructions stored in the memory 1510.

[0123] The travel control device 1500 can further comprise an input / output interface 1530, a network interface 1540, a storage interface 1550, etc. These interfaces 1530, 1540, 1550 and the memory 1510 and the processor 1520 can be connected through a bus 1560, for example.

[0124] The memory 1510 can comprise a system memory, a fixed non-volatile storage medium, etc., for example. The system memory stores, for example, an operating system, an application program, a Boot Loader, and other programs, etc.

[0125] The processor 1520 can be implemented with a general-purpose processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other programmable logic device, a discrete gate or transistor logic, or the like discrete hardware component.

[0126] The input / output interface 1530 provides a connection interface for input / output devices such as a display, a mouse, a keyboard, a touch screen, and the like. The network interface 1540 provides a connection interface for various networking devices. The storage interface 1550 provides a connection interface for external storage devices such as an SD card, a U disk, and the like. The bus 1560 can use any of a variety of bus structures. For example, the bus structure includes, but is not limited to, an Industry Standard Architecture (ISA) bus, a Micro Channel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus.

[0127] The driving control scheme of various embodiments of the present disclosure can be applied to new energy vehicles or other intelligent driving vehicles, for example, for emergency braking decision under new energy driving modes such as plug-in hybrid driving, pure electric driving, and fuel cell driving.

[0128] Those skilled in the art should understand that the embodiments of the present disclosure can be provided as a method, a system, or a computer program product. Therefore, the present disclosure can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present disclosure can take the form of a computer program product implemented on one or more (non-transitory) computer-readable storage media (including but not limited to disk storage, CD-ROM, optical storage, cloud storage, etc.) containing computer program code. The computer program product should be understood as a software product that mainly realizes its solutions through a computer program.

[0129] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart or flows and / or block or blocks in the block diagrams. Figure 1 one or more functions specified in the flowchart or flows and / or block or blocks of the block diagrams. Figure 1 one or more functions specified in the flowchart or flows and / or block or blocks of the block diagrams.

[0130] These computer program instructions can also be stored in a computer readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer readable memory produce an article of manufacture including instructions which implement the Figure 1 one or more functions specified in the flowchart or flows and / or block or blocks of the block diagrams. Figure 1 one or more functions specified in the flowchart or flows and / or block or blocks of the block diagrams.

[0131] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart or flows and / or block or blocks in the block diagrams. Figure 1 one or more functions specified in the flowchart or flows and / or block or blocks of the block diagrams. Figure 1 one or more functions specified in the flowchart or flows and / or block or blocks of the block diagrams.

[0132] The above description is only preferred embodiments of the present disclosure, not intended to limit the present disclosure. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

1. A driving control method, characterized in that, include: In response to an emergency braking decision, determine the trajectory points of the intelligent driving vehicle and the target that triggered the emergency braking decision within a specified future time period; Based on the trajectory points of the autonomous vehicle and the target that triggered the emergency braking decision within a specified future time period, determine whether there is a risk of conflict between the target and the autonomous vehicle, including: When the radial components of the target's and the autonomous vehicle's velocities are in the same direction, if the velocities of the autonomous vehicle and the target will not reach the same point at any time during the deceleration process of the autonomous vehicle at its maximum braking deceleration, the distance traveled by the target from its current radial velocity component to zero according to the radial acceleration component is determined as the target's braking distance. If the velocities of the autonomous vehicle and the target reach the same point at any time during the deceleration process of the autonomous vehicle at its maximum braking deceleration, the distance traveled by the target from its current radial velocity component to the point where the speeds are the same is determined as the target's braking distance. By comparing the difference between the braking distance of the autonomous vehicle and the braking distance of the target with the radial distance between the target and the autonomous vehicle, it is determined whether there is a risk of conflict between the target and the autonomous vehicle in the radial direction. The radial distance between the target and the autonomous vehicle is determined based on the radial component of the target's position in the autonomous vehicle's coordinate system. or / and, Based on the radial distance between the target and the autonomous vehicle, and the difference between the radial components of the autonomous vehicle's velocity and the target's velocity, the time when the target and the autonomous vehicle collide in the radial direction is determined. Based on the target's current lateral position component, velocity lateral component, and acceleration lateral component in the autonomous vehicle's coordinate system, and the time of the collision, the lateral position component of the target at the time of the collision is determined. Based on the lateral position component of the target at the time of the collision, it is determined whether the target will move into the width range of the autonomous vehicle in the lateral direction. Based on whether the target will move into the width range of the autonomous vehicle in the lateral direction, it is determined whether there is a risk of collision between the target and the autonomous vehicle in the lateral direction. If a risk of conflict is identified, maintain the emergency braking decision; if no risk of conflict is identified, cancel the emergency braking decision.

2. The driving control method according to claim 1, characterized in that, Based on the trajectory points of the autonomous vehicle and the target that triggers the emergency braking decision over a specified future time period, determine whether there is a risk of conflict between the target and the autonomous vehicle, including: If the trajectory point of the target that triggers the emergency braking decision is located within the vehicle frame of the intelligent driving vehicle at any time within a specified future time period, it is determined that there is a risk of conflict between the target and the intelligent driving vehicle.

3. The driving control method according to claim 1, characterized in that, Based on the trajectory points of the autonomous vehicle and the target that triggers the emergency braking decision over a specified future time period, determine whether there is a risk of conflict between the target and the autonomous vehicle, including: If the line connecting the trajectory points of the target that triggers the emergency braking decision intersects with the line connecting the trajectory points of the autonomous vehicle within the same specified time period, it is determined that there is a risk of conflict between the target and the autonomous vehicle.

4. The driving control method according to claim 1, characterized in that, Based on the trajectory points of the autonomous vehicle and the target that triggers the emergency braking decision over a specified future time period, determine whether there is a risk of conflict between the target and the autonomous vehicle, including: Determine whether the trajectory point of the target that triggers the emergency braking decision is within the vehicle frame of the intelligent driving vehicle at any time within a specified future time period. If the judgment result is yes, it is determined that there is a risk of conflict between the target and the intelligent driving vehicle. If the judgment result is negative, it is determined whether the line connecting the trajectory points of the target that triggered the emergency braking decision in the future intersects with the line connecting the trajectory points of the intelligent driving vehicle in the future specified time period. If the judgment result is that they intersect, it is determined that there is a risk of conflict between the target and the intelligent driving vehicle.

5. The driving control method according to any one of claims 1-4, characterized in that, Based on the trajectory points of the autonomous vehicle and the target that triggers the emergency braking decision over a specified future time period, determine whether there is a risk of conflict between the target and the autonomous vehicle, including: If no conflict risk is determined based on the trajectory points, the emergency braking decision is made again based on the motion state information of the last trajectory point of the intelligent driving vehicle and the target that triggers the emergency braking decision within a specified time period in the future. If the assessment result is still that emergency braking is required, it is determined that there is a risk of conflict between the target and the autonomous vehicle. If the assessment result is that emergency braking is not required, it is determined that there is no risk of conflict between the target and the autonomous vehicle.

6. The driving control method according to claim 5, characterized in that, The determination to apply emergency braking again includes: Based on the target's position, velocity, and acceleration in the autonomous vehicle's coordinate system at the last trajectory point, as well as the autonomous vehicle's speed and braking attributes, determine whether there is a risk of conflict between the target and the autonomous vehicle. If a risk of conflict is identified, it is determined that emergency braking is necessary.

7. The driving control method according to claim 6, wherein, Determining whether there is a conflict risk between the target and the autonomous vehicle includes: The relative speed of the target with respect to the autonomous vehicle at the last trajectory point is determined, and whether the target is a candidate target for triggering emergency braking is determined based on whether the ray of the target in the direction of relative speed crosses the safe driving range of the autonomous vehicle. When the target is a candidate target that could trigger emergency braking, the system determines whether there is a risk of conflict between the target and the autonomous vehicle based on the target's position, velocity, and acceleration in the autonomous vehicle's coordinate system at the last trajectory point, as well as the autonomous vehicle's speed and braking attributes.

8. A driving control device, comprising: Memory; And a processor coupled to the memory, characterized in that the processor is configured to execute the driving control method of any one of claims 1-7 based on instructions stored in the memory.

9. An intelligent driving vehicle, characterized in that, include: A driving control device is configured to perform the driving control method according to any one of claims 1-7.

10. A computer-readable storage medium storing computer instructions thereon, characterized in that, When the computer instructions are executed by the processor, they implement the steps of the driving control method according to any one of claims 1-7.

11. A computer program product comprising computer instructions, characterized in that, When the computer instructions are executed by the processor, they implement the steps of the driving control method according to any one of claims 1-7.

Images