Obstacle avoidance control method, device and equipment for vehicle

Abstract

The invention provides an obstacle avoidance control method, device and equipment for a vehicle, and is applied to the technical field of intelligent driving. The method comprises the steps that vehicle state information, driver state information and environment information of a vehicle at the current moment are collected; the environment information comprises target information of a target object located in the vehicle blind area. According to the vehicle state information and the driver state information, the driving intention of the driver at the next moment and multiple pieces of alternative driving information of the vehicle at the next moment are predicted; and performing space-time coupling analysis on the target information and each piece of alternative driving information, and predicting risk information corresponding to each piece of alternative driving information. And when it is detected that a risk coupling point exists in the risk information, obstacle avoidance information is generated according to the vehicle state information, the environment information and the risk information so as to avoid the target object. The problem that a traditional blind area monitoring system is low in monitoring reliability due to passivity, lag and intention misjudgment can be solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent driving, in particular to a vehicle obstacle avoidance control method, device and equipment. BACKGROUND

[0002] In the field of active safety of automobiles, blind area monitoring system has become an important auxiliary function, and its core task is to detect potential obstacles in the blind area of the driver's field of view and issue a warning when a risk occurs. However, the working mode of the existing blind area detection system is essentially a passive and instantaneous state-based response mechanism, and the system usually triggers an alarm only when the target enters the blind area of the vehicle or reaches a certain critical distance from the vehicle.

[0003] However, this passive response mechanism has some problems in actual work. Specifically, when the driver intends to change lanes or turn, if there is a high-speed approaching vehicle in the blind area, the system often issues an alarm just before the conflict occurs, which can reduce the driver's reaction time and cause accidents. Secondly, during normal driving when the driver has no intention to change lanes, the system will still alarm harmless vehicles in the blind area, and this frequent false alarm will seriously undermine the driver's trust in the system, leading him to ignore the warning when real danger occurs.

[0004] Therefore, there is an urgent need for a vehicle obstacle avoidance control method, device and equipment to solve the problem of low monitoring reliability caused by passive, lagging and intention misjudgment of the traditional blind area monitoring system. SUMMARY

[0005] The purpose of the present application is to provide a vehicle obstacle avoidance control method, device and equipment to solve the problem of low monitoring reliability caused by passive, lagging and intention misjudgment of the traditional blind area monitoring system.

[0006] In a first aspect, the embodiments of the present application provide a vehicle obstacle avoidance control method, which comprises: collecting vehicle state information, driver state information and environment information of the vehicle at the current time. The environment information includes target information of a target object located in the blind area of the vehicle. According to the vehicle state information and the driver state information, the driving intention of the driver at the next time and the multiple alternative driving information of the vehicle at the next time are predicted. The target information is coupled with each alternative driving information for space-time analysis, and the risk information corresponding to each alternative driving information is predicted. In the case where a risk coupling point is detected in the risk information, the avoidance information is generated according to the vehicle state information, the environment information and the risk information to avoid the target object.

[0007] The obstacle avoidance control method for vehicles provided in this application deeply integrates vehicle status, driver's real-time operational intentions, and blind spot environmental information to predict the driver's short-term operational intentions at the next moment, and generates multiple probabilistic alternative driving trajectories accordingly. Furthermore, by performing high-precision spatiotemporal coupling analysis between the motion trajectory of blind spot targets and the vehicle's possible paths, high-risk coupling points can be identified before potential conflicts occur, thereby achieving early warning of risks. Simultaneously, by combining real-time vehicle dynamics and surrounding environmental constraints, this application proactively generates smooth, feasible, and driver-intention-compliant collaborative obstacle avoidance paths or operational guidance, significantly extending the driver's decision-making and reaction window, effectively avoiding safety accidents caused by delayed or false alarms, and increasing the driver's trust in intelligent driving.

[0008] One possible implementation involves predicting the driver's driving intention at the next moment and multiple alternative driving options for the vehicle at the next moment, based on vehicle state information and driver state information. This includes: predicting the driver's driving intention at the next moment based on vehicle state information and driver state information; and predicting multiple alternative driving trajectories for the vehicle at the next moment and the probability of driving each alternative driving trajectory based on the driving intention and a pre-built vehicle dynamics model.

[0009] One possible implementation involves performing spatiotemporal coupling analysis between the target information and each alternative driving information to predict the risk information corresponding to each alternative driving information. This includes: predicting the target trajectory of the target object at the next moment based on the target information; and for any alternative driving information among all alternative driving information, performing spatiotemporal coupling analysis between the target trajectory and the alternative driving trajectory of the target alternative driving information to predict the risk index of the vehicle colliding with the target object at the next moment and the time of the collision.

[0010] One possible implementation, the vehicle obstacle avoidance control method provided in this application embodiment, further includes: when the risk index of the vehicle colliding with the target object at the next moment exceeds the preset risk index, and the time of the vehicle colliding with the target at the next moment is less than the safe time, determining that there is a risk coupling point in the risk information.

[0011] One possible implementation involves generating obstacle avoidance information based on vehicle status information, environmental information, and risk information. This includes: determining a guidance plan for the vehicle based on the environmental and risk information; and determining corresponding prompts based on the guidance plan, including the vehicle status information, target information, and risk information.

[0012] One possible implementation involves determining a vehicle guidance plan based on environmental and risk information, including: modifying alternative driving paths corresponding to risk coupling points in the risk information based on environmental information to determine the vehicle's obstacle avoidance path in the next moment; and / or, determining operational suggestions for the driver in the next moment based on environmental and risk information; and / or, if the vehicle's auxiliary control function is detected to be activated, generating auxiliary control commands for the vehicle based on environmental and risk information.

[0013] One possible implementation involves determining the corresponding prompts for the guidance plan based on vehicle status information, target information, and risk information. This includes: determining the relative speed between the vehicle and the target object based on the vehicle status information and target information; determining the vehicle's risk level based on the relative speed, the probability of driving the alternative driving path corresponding to the risk coupling point in the alternative driving information, and the time when the vehicle will collide with the target object in the next moment, as indicated in the risk information; and determining the corresponding prompts for the guidance plan based on the risk level.

[0014] Secondly, this application provides an obstacle avoidance control device for a vehicle, which includes a data acquisition module, a prediction module, and a generation module.

[0015] The data acquisition module is used to collect vehicle status information, driver status information, and environmental information at the current moment. Environmental information includes target information of objects located in the vehicle's blind spots.

[0016] The prediction module is used to predict the driver's driving intention at the next moment and multiple alternative driving information for the vehicle at the next moment, based on vehicle status information and driver status information.

[0017] The prediction module is also used to perform spatiotemporal coupling analysis between the target information and each alternative driving information to predict the risk information corresponding to each alternative driving information.

[0018] The generation module is used to generate obstacle avoidance information based on vehicle status information, environmental information, and risk information when a risk coupling point is detected in the risk information, so as to avoid the target object.

[0019] One possible implementation involves the prediction module, when predicting the driver's driving intention and multiple alternative driving options for the vehicle at the next moment based on vehicle and driver state information, specifically predicting the driver's driving intention based on the vehicle and driver state information. Based on the driving intention and a pre-built vehicle dynamics model, it predicts multiple alternative driving trajectories for the vehicle at the next moment, as well as the probability of driving each alternative trajectory.

[0020] One possible implementation involves the prediction module performing spatiotemporal coupling analysis between the target information and each alternative driving information to predict the risk information corresponding to each alternative driving information. Specifically, this involves predicting the target trajectory of the target object at the next moment based on the target information. For any alternative driving information among all the alternative driving information, the module performs spatiotemporal coupling analysis between the target trajectory and the alternative driving trajectory of the target information to predict the risk index of a collision between the vehicle and the target object at the next moment and the time of the collision.

[0021] In one possible implementation, the obstacle avoidance control device for vehicles provided in this application embodiment is further used to determine that there is a risk coupling point in the risk information when the risk index of the vehicle colliding with the target object at the next moment exceeds the preset risk index, and the time of the vehicle colliding with the target at the next moment is less than the safe time.

[0022] One possible implementation is that, when the generation module generates obstacle avoidance information based on vehicle status information, environmental information, and risk information, it specifically determines the vehicle guidance plan based on the environmental and risk information. Based on the vehicle status information, target information, and risk information, it determines the corresponding prompts for the guidance plan.

[0023] One possible implementation is that, when determining the vehicle guidance plan based on environmental and risk information, the generation module specifically modifies the alternative driving paths corresponding to the risk coupling points in the risk information based on the environmental information to determine the obstacle avoidance path for the vehicle in the next moment. And / or, based on the environmental and risk information, it determines operational suggestions for the driver in the next moment. And / or, when the vehicle's auxiliary control function is detected to be activated, it generates auxiliary control commands for the vehicle based on the environmental and risk information.

[0024] One possible implementation involves the generation module, when determining the prompt information corresponding to the guidance plan based on vehicle status information, target information, and risk information, specifically determining the relative speed between the vehicle and the target object based on the vehicle status information and target information. Based on the relative speed, the probability of driving the alternative driving path corresponding to the risk coupling point in the alternative driving information, and the time when the vehicle will collide with the target object in the next moment as indicated in the risk information, the vehicle's risk level is determined. Based on the risk level, the prompt information corresponding to the guidance plan is determined.

[0025] Thirdly, embodiments of this application provide a vehicle obstacle avoidance control device that has the function of implementing the vehicle obstacle avoidance control method of the first aspect or any possible implementation thereof. This function can be implemented by hardware or by hardware executing corresponding software. The hardware or software includes one or more modules corresponding to the above-described function.

[0026] Fourthly, embodiments of this application provide a computer-readable storage medium storing instructions that, when executed on a computer, enable the computer to perform the obstacle avoidance control method for a vehicle according to the first aspect or any possible implementation thereof.

[0027] Fifthly, embodiments of this application provide a computer program product containing instructions that, when run on a computer, enable the computer to execute the obstacle avoidance control method for a vehicle as described in the first aspect or any possible implementation thereof.

[0028] The technical effects of any of the design methods in aspects two through five can be found in aspect one or in different possible implementations of aspect one, and will not be repeated here. Attached Figure Description

[0029] To more clearly illustrate the technical solutions in the specific embodiments of this application or the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0030] Figure 1 A system structure diagram of a vehicle obstacle avoidance control system provided in an embodiment of this application; Figure 2 A specific example diagram of a vehicle obstacle avoidance control system provided in an embodiment of this application; Figure 3 A flowchart illustrating a vehicle obstacle avoidance control method provided in this application embodiment; Figure 4 A specific example diagram of a vehicle obstacle avoidance control method provided in this application embodiment; Figure 5 A schematic diagram of a vehicle obstacle avoidance control device provided in an embodiment of this application; Figure 6 This is another system architecture diagram of a vehicle obstacle avoidance control system provided in an embodiment of this application. Detailed Implementation

[0031] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0032] Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of the application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0033] During vehicle operation, the inherent limitations of the vehicle body structure (such as A / B / C pillars) and the field of vision of the rearview mirrors create unavoidable blind spots. Especially during critical maneuvers such as lane changes and turns, if other vehicles, pedestrians, or non-motorized vehicles lurking in the blind spots are not detected in time, collisions can easily occur, posing a serious threat to road safety.

[0034] To address this potential hazard, relevant technologies primarily rely on various sensors for environmental perception. However, single-sensor solutions each have their limitations. For example, ultrasonic radar has a short detection range and is susceptible to environmental interference, making it unsuitable for high-speed scenarios; millimeter-wave radar, while accurate in speed measurement and possessing strong penetration, suffers from low angular resolution and weak ability to identify static targets; cameras provide rich visual information, but their performance degrades sharply in strong light, at night, or in adverse weather conditions. Some improved solutions attempt to integrate multiple sensors, such as radar and cameras working together. However, the principle of this collaboration often relies on fixed rules for data fusion, lacking the ability to dynamically adjust weights based on environmental changes. This leads to reduced system reliability when sensors are interfered with, and the risks of false alarms and missed alarms remain difficult to avoid.

[0035] Based on this, embodiments of this application provide a vehicle obstacle avoidance control method. This method includes collecting vehicle state information, driver state information, and environmental information at the current moment. The environmental information includes target information of a target object located in the vehicle's blind spot. Based on the vehicle state information and driver state information, the driver's driving intention at the next moment and multiple alternative driving information for the vehicle at the next moment are predicted. The target information is spatiotemporally coupled with each alternative driving information to predict the risk information corresponding to each alternative driving information. If a risk coupling point is detected in the risk information, obstacle avoidance information is generated based on the vehicle state information, environmental information, and risk information to avoid the target object.

[0036] The obstacle avoidance control method for vehicles provided in this application deeply integrates vehicle status, driver's real-time operational intentions, and blind spot environmental information to predict the driver's short-term operational intentions at the next moment, and generates multiple probabilistic alternative driving trajectories accordingly. Furthermore, by performing high-precision spatiotemporal coupling analysis between the motion trajectory of blind spot targets and the vehicle's possible paths, high-risk coupling points can be identified before potential conflicts occur, thereby achieving early warning of risks. Simultaneously, by combining real-time vehicle dynamics and surrounding environmental constraints, this application proactively generates smooth, feasible, and driver-intention-compliant collaborative obstacle avoidance paths or operational guidance, significantly extending the driver's decision-making and reaction window, effectively avoiding safety accidents caused by delayed or false alarms, and increasing the driver's trust in intelligent driving.

[0037] The methods provided in the embodiments of this application will now be described in conjunction with the specific accompanying drawings.

[0038] On the one hand, embodiments of this application provide a vehicle obstacle avoidance control system. For example... Figure 1 As shown, the obstacle avoidance control system 100 of the vehicle may include: a sensor module 101, a control module 102, and an interaction module 103.

[0039] The sensor module 101 is used to collect information about the vehicle's internal and external environment, including vehicle status information, driver information, and environmental information. The sensor module 11 may include a camera, millimeter-wave radar, and a V2X antenna, among other things.

[0040] For example, such as Figure 2As shown, a forward-facing camera (e.g., position A) is mounted on the top of the vehicle's windshield. This camera acquires environmental information about the road ahead and has a resolution of 1280x720 at 30fps and a field of view of 120°. Side-view cameras (e.g., positions B and C) are deployed below the left and right rearview mirrors, respectively. These cameras acquire environmental information about a 120° blind spot to the side and rear of the vehicle, as well as target information about objects within that blind spot. These side-view cameras feature automatic defogging and low-light enhancement functions, and also have a resolution of 1280x720 at 30fps and a field of view of 120°. A driver monitoring camera (e.g., position F) is located above the steering wheel inside the vehicle. This camera tracks the driver's gaze direction and head posture (tilt / yaw angle) in real time with an accuracy of ±3°. This driver monitoring camera can use a near-infrared lens. Millimeter-wave radars (such as at positions D and E) can be deployed on the left and right sides of the vehicle's rear bumper. These radars can be used to acquire target information of objects located in the vehicle's blind spots. The detection range of these millimeter-wave radars is 0.2-150 meters, and the angular coverage is ±60°. A V2X antenna (such as at position H) can be deployed at the rear of the vehicle's roof. This V2X antenna is used to exchange data with other intelligent connected vehicles and roadside infrastructure (such as intelligent traffic lights and roadside units) in the vicinity.

[0041] The control module 102, after receiving vehicle status information, driver status information, and environmental information uploaded by the sensor module 101, uses the vehicle obstacle avoidance control method provided in this application to predict the driver's short-term operational intentions based on the vehicle status information and driver status information, and generates multiple future driving trajectories with probability distributions. It then performs spatiotemporal coupling analysis on the predicted trajectories of blind spot targets and the vehicle's alternative trajectories to identify high-risk coupling points. Finally, combining vehicle dynamics constraints and the real-time environment, it plans the optimal cooperative obstacle avoidance path or operational strategy to assist the driver in avoiding target objects located in the vehicle's blind spot.

[0042] The interaction module 103, under the control of the control module 102, converts the obstacle avoidance information generated by the control module 102 into intuitive guidance information. This module employs an augmented reality head-up display system, such as an AR-HUD projector located above the dashboard (e.g.,...). Figure 2 At position G (as shown), the projector can project at a distance of 7.5 meters and a field of view of 10° × 5°, with a resolution exceeding 12000 cd / m². 2The brightness of the light will overlay the safety guide lines onto the vehicle's dashboard or real-world road scene. Simultaneously, the interactive module 103 is equipped with a multi-layered haptic feedback system, including a steering wheel torque motor and a seat zone vibration motor. Combined with in-vehicle speakers supporting 3D directional alerts, this forms a multi-channel warning system of vision, touch, and hearing, ensuring accurate and timely obstacle avoidance guidance for the driver at different risk levels.

[0043] It should be noted that the above Figure 1 The obstacle avoidance control system 100 of the vehicle shown is merely an example of the application scenario of the solution in this application, and is not intended to limit the application scenario of the solution in this application.

[0044] On the one hand, embodiments of this application provide a vehicle obstacle avoidance control method, which can be implemented by... Figure 1 The obstacle avoidance control system 100 of the vehicle shown is activated. For example... Figure 3 As shown, the method may include the following steps.

[0045] S301 collects vehicle status information, driver status information, and environmental information at the current moment.

[0046] One possible implementation is to obtain the vehicle's current status information in real time via the vehicle's CAN bus.

[0047] The vehicle status information may include the vehicle's motion parameters, driver input, and vehicle signal status. For example, vehicle motion parameters may include real-time vehicle speed, longitudinal and lateral acceleration, and yaw rate. Driver input may include steering wheel angle and its rate of change (steering angular velocity), and the opening of the accelerator and brake pedals and their rates of change. Vehicle signal status may include the on / off status of the left and right turn signals.

[0048] Specifically, the system reads the corresponding data frames of vehicle status information from the CAN bus. The data in these frames is uploaded by the vehicle's wheel speed sensors, inertial measurement unit (IMU), steering angle sensor, pedal travel sensor, and body controller.

[0049] Another possible approach is to obtain driver status information from inside the vehicle using the vehicle's sensors.

[0050] Specifically, the Driver Monitoring System (DMS) camera uses computer vision algorithms to calculate the driver's 3D gaze direction, determining whether their gaze is focused on key areas such as the left / right side mirrors, outside the side windows, or the road ahead. Based on the DMS camera, the driver's head yaw and pitch angles are estimated to determine whether their attention is focused on the blind spot. Sensors built into the steering wheel or seat (such as photoplethysmography (PPG) sensors and skin conductance response (GSR) sensors) monitor the driver's heart rate, heart rate variability, and skin conductance responses to help assess their current level of stress or workload.

[0051] For example, 68 key points on the driver's face are extracted by calling the OpenFace library, and the 3D gaze vector is calculated using the PnP algorithm. This vector is then mapped to a preset area of ​​interest (such as the left / right rearview mirrors or the road ahead) to determine the driver's gaze direction. Based on the Dlib library, the solvePnP function is used to solve the problem, ultimately outputting the yaw and pitch angles that represent the head orientation.

[0052] Another possible implementation is to obtain the vehicle's environmental information through millimeter-wave radar installed on the left and right sides of the rear bumper and side-view cameras on the left and right exterior rearview mirrors.

[0053] The environmental information may include current lane information, adjacent lane information, the status of vehicles in front and behind, and target information of objects located in the vehicle's blind spot. This target information may include the distance, relative speed, and azimuth angle between the target object and the vehicle, as well as the type, size, and predicted trajectory of the target object.

[0054] For example, the raw point cloud data acquired by millimeter-wave radar is processed using the DBSCAN clustering algorithm (parameters set to eps=0.5 m, min_samples=3) to filter out noise and effectively separate independent targets, ultimately outputting the target's position (x, y), velocity (vx, vy), and radar cross-section value. Based on the deployed lightweight YOLOv7-tiny target detection model (input size fixed at 640x384), the bounding box, specific category (e.g., vehicle, pedestrian, bicycle), and confidence score of target objects in the images acquired by the camera are output in real time. The detected target objects are tracked across frames using a Kalman filter algorithm to maintain the stability of their target IDs.

[0055] S302 predicts the driver's driving intention at the next moment and multiple alternative driving information for the vehicle at the next moment based on vehicle status information and driver status information.

[0056] The alternative driving information includes alternative driving trajectories and the probability of driving the alternative driving trajectories.

[0057] Specifically, firstly, based on vehicle and driver status information, the driver's driving intentions in the near future are predicted. Then, based on the predicted driving intentions and combined with a high-precision vehicle dynamics model, multiple possible driving trajectories of the vehicle over a period of time are deduced, and a corresponding probability of occurrence is evaluated for each trajectory.

[0058] One possible implementation involves predicting the driver's driving intention at the next moment based on vehicle and driver state information. Then, based on the driving intention and a pre-built vehicle dynamics model, multiple alternative driving trajectories for the vehicle at the next moment, along with the probability of driving each alternative trajectory, are predicted.

[0059] For example, a standardized input feature vector is constructed using real-time acquired vehicle and driver status information at a fixed time period (e.g., 200 milliseconds). This feature vector integrates the vehicle's direct control signals and the driver's attention information, specifically including: steering wheel angle θ, steering angular velocity dθ / dt, throttle opening α, brake opening β, left / right turn signal status (represented by a Boolean value of 0 or 1), whether the driver's gaze is on the left / right rearview mirror (represented by a Boolean value of 0 or 1), and head yaw angle.

[0060] Feature vectors are input into a pre-trained deep learning model for real-time inference. This prediction model can be a two-layer long short-term memory network with 128 neurons in its hidden layer. This model effectively captures the time-series dependencies and dynamic patterns of driver actions. The deep learning model processes the input feature vectors to obtain the probability distribution of five typical driving intentions within a predetermined time window (e.g., the next 3 seconds). These intention classifications can include: going straight, preparing to change lanes to the left, preparing to change lanes to the left, preparing to change lanes to the right, and preparing to change lanes to the right. The characteristics of preparing to change lanes to the left or right are typically manifested by the absence of a turn signal but with eye contact / head movement towards the corresponding side mirror, accompanied by a slight turn of the steering wheel. The characteristics of changing lanes to the left / right are typically manifested by the presence or absence of a turn signal but with clear steering wheel input. The characteristics of preparing to turn left / right are typically determined in intersection scenarios by combining corresponding vehicle deceleration, turn signal operation, and eye contact features. The characteristics of a clear acceleration / deceleration intention are determined based on the depth and rate of change of the accelerator / brake pedal.

[0061] Furthermore, based on the predicted multiple possible driving intentions (with a particular focus on high-risk intention categories such as lane changes), combined with a high-precision vehicle dynamics model, parallel and real-time trajectory extrapolation is performed in a virtual simulation environment to generate multiple possible driving trajectories for the vehicle in the future (e.g., 3 to 5 seconds), and each trajectory is assigned a driving probability.

[0062] The vehicle dynamics model is a high-fidelity mathematical model that fully considers the vehicle's physical characteristics and its interaction with the environment. Modeling parameters include, but are not limited to: vehicle mass, suspension characteristics, tire models (such as the "magic formula" Pacejka model), current real-time vehicle speed, and estimated road adhesion coefficient. The dynamic parameters of the Pacejka model are as follows.

[0063]

[0064] in, and These are the vehicle's longitudinal and lateral speeds, respectively. The yaw rate is angular velocity. This refers to the steering angle of the front wheels. and These are the longitudinal forces on the front and rear wheels, respectively. and To separate the lateral forces on the front and rear wheels, these tire forces are calculated using the "magic formula" tire model based on the tire's slip ratio and sideslip angle. Model parameters include the vehicle's mass *m* and the drag coefficient. Distance from the front and rear axles to the center of gravity and and the vehicle's moment of inertia about its vertical axis. .

[0065] The trajectory prediction process specifically considers the uncertainty of driver actions and the randomness of the vehicle's dynamic response. Therefore, for the same driving intention, the system generates multiple trajectory hypotheses with different styles to cover various possibilities. For example, when the predicted intention is "change lanes to the left," the system will extrapolate at least three typical trajectory hypotheses in parallel: 1) Aggressive lane change trajectory. This aggressive lane change trajectory corresponds to the condition where the driver applies a large steering wheel angle and a high steering angular velocity, resulting in rapid changes in the vehicle's lateral displacement. For example, a step input of the steering wheel to the target angle (e.g., 30°), a steering angular velocity (e.g., >5° / s), and DMS line-of-sight confirmation.

[0066] 2) Smooth lane change trajectory. This smooth lane change trajectory corresponds to the driver applying small but stable steering wheel input, with the vehicle completing the lane change in a relatively gentle manner. For example, steering wheel angle (e.g., 15° to 30°) and steering angular velocity (e.g., 2° to 5° / s).

[0067] 3) Abandoning the lane change trajectory. This refers to the driver initiating a lane change intention but ceasing the operation due to changes in the environment, straightening the steering wheel, and ultimately keeping the vehicle in the original lane.

[0068] Each trajectory hypothesis is assigned an initial probability based on its corresponding driver operation characteristics (such as the magnitude of steering angular velocity) and the statistical frequency of this operation style in historical data. Finally, a trajectory set {T} containing the predicted vehicle pose (coordinates, heading angle) and its corresponding probabilities for future time series is output. i , P i}, where P i Represents trajectory T i The likelihood of it happening.

[0069] S303 performs spatiotemporal coupling analysis between the target information and each alternative driving information to predict the risk information corresponding to each alternative driving information.

[0070] The risk information includes the probability of the vehicle colliding with the target in the next moment or the time when the vehicle will collide with the target in the next moment.

[0071] Specifically, based on the target information obtained, the target trajectory of the target object at the next moment is predicted.

[0072] For example, a constant rotation and acceleration model (CTRA) is used as the basic predictive model for the target motion. The model is as follows.

[0073]

[0074] Where x is the x-coordinate of the target object in two-dimensional space, y is the y-coordinate of the target object in two-dimensional space, v is the velocity of the target object in two-dimensional space, and θ is the heading angle of the target object in two-dimensional space. Δt represents the yaw rate (rotation rate) of the target object in two-dimensional space. Δt is the time step a. k Acceleration for the target.

[0075] Then, for any target alternative driving information among all alternative driving information, a spatiotemporal coupling analysis is performed between the target trajectory and the target alternative driving trajectory of the target alternative driving information to predict the probability of the vehicle colliding with the target object at the next moment and the time of the collision between the vehicle and the target at the next moment.

[0076] For example, such as Figure 4 As shown, for each alternative trajectory T of the vehicle i For each blind spot target's predicted trajectory, a pairwise spatiotemporal conflict detection is performed to determine the minimum distance between the vehicle and the target object and the collision time.

[0077] The minimum distance can be obtained by determining the minimum Euclidean distance between corresponding points on the two trajectories within the simulation time window. The collision time can be the time required for the distance between corresponding points on the two trajectories to first fall below the safe radius.

[0078] The safety radius is determined by the following formula.

[0079]

[0080] in, The safe radius between the vehicle and the target object. For the width of the vehicle, 0.5 represents the width of the target object, with a safety margin of 0.5.

[0081] The collision time can be determined using the following formula.

[0082]

[0083] in, For the collision time, For the safety radius.

[0084] Furthermore, in order to comprehensively assess the severity of the risk, this application will also calculate a risk index.

[0085] Specifically, the probability P of the vehicle trajectory occurring... i Time urgency (due to collision time) Through the exponential decay function Characterization, in which (as the decay constant), and the severity of potential collisions (via relative velocity). (Weighted calculation). The formula for calculating this risk index is as follows.

[0086]

[0087] in, This is a risk index.

[0088] If the probability of the vehicle colliding with the target object in the next moment exceeds the preset probability, and the time between the vehicle colliding with the target in the next moment is less than the safe time, then a risk coupling point is determined to exist in the risk information.

[0089] For example, when a risk index is detected >0.8, and collision time If the time is less than 2.5 seconds, a risk coupling point is identified.

[0090] S304: When a risk coupling point is detected in the risk information, obstacle avoidance information is generated based on vehicle status information, environmental information, and risk information to avoid the target object.

[0091] Specifically, based on environmental and risk information, a vehicle guidance plan is determined. Based on vehicle status information, target information, and risk information, corresponding prompts are determined for the guidance plan.

[0092] The guidance scheme may include the vehicle's obstacle avoidance path in the next moment, operational suggestions for the driver in the next moment, and auxiliary control commands for the vehicle.

[0093] One possible approach is to modify the alternative driving paths corresponding to the risk coupling points in the risk information based on environmental information to determine the obstacle avoidance path of the vehicle in the next moment.

[0094] Specifically, the alternative driving paths corresponding to the driver's driving intentions are revised to obtain the obstacle avoidance path for the vehicle in the next moment. This obstacle avoidance path can safely avoid target objects located in the vehicle's blind spot. The revision strategies include, but are not limited to, initiating a lane change slightly later to give way to a high-speed approaching target, using a gentler lane change angle to increase the safety margin between the vehicle and the target object, or creating a safe lane change window through precise acceleration / deceleration. This optimized trajectory can be overlaid on the driver's real field of vision as a visual guide line through AR-HUD or digital dashboard, providing the driver with intuitive path guidance.

[0095] The obstacle avoidance path can be determined using the following path optimization model.

[0096]

[0097] The constraints of this path optimization model are as follows: , , as well as . Let 'a' be the front wheel steering angle and 'a' be the acceleration. For reference trajectory, This is a vehicle dynamics model.

[0098] Another possible approach is to determine the driver's operational recommendations for the next moment based on environmental and risk information.

[0099] Specifically, if the system detects that the driver's intended trajectory poses a high risk that cannot be effectively avoided through fine-tuning (e.g., a target in the blind spot is traveling too fast and is extremely close), it will generate alternative operational suggestions. For example, it may suggest that the driver abandon the lane change and maintain the current lane, or, if the environment permits, prompt the driver to move to a safe area on the other side to avoid the target. At the same time, the system will also provide precise speed adjustment suggestions, clearly indicating the timing and magnitude of acceleration or deceleration, to create a safe space in conjunction with path planning.

[0100] Another possible implementation is to generate auxiliary control commands for the vehicle based on environmental and risk information when the vehicle's auxiliary control function is detected to be activated.

[0101] Specifically, when the driver has authorized the system (e.g., by activating a specific advanced driver assistance mode) or in an emergency, the system can generate auxiliary control commands to subtly intervene in the vehicle to assist the driver. For example, it can apply a slight compensating torque through the electric power steering system to guide the vehicle along a system-recommended safe trajectory. Alternatively, it can use the ESP system or engine management system to apply slight automatic braking or acceleration to adjust the vehicle speed to match a safe path.

[0102] This approach prioritizes guidance and collaboration, with strong intervention as a secondary measure, while respecting the driver's final decision-making authority.

[0103] Furthermore, based on vehicle status information and target information, the relative speed between the vehicle and the target object is determined. Based on the relative speed, the probability of driving the alternative driving path corresponding to the high-risk coupling point in the alternative driving information, and the time when the vehicle will collide with the target object in the next moment, the vehicle's risk level is determined. Based on the risk level, the corresponding prompts for the guidance plan are determined.

[0104] Specifically, based on the real-time calculated risk level, the intensity and form of the prompts are dynamically adjusted to achieve contextualized warnings and guidance. For example, when the risk index is less than 0.4, the risk level can be determined as no risk, and the system remains silent, without providing any prompts to avoid interfering with the driver. When the risk index is greater than 0.4 and less than 0.6, the risk level can be determined as low risk, and a blind spot target icon (such as yellow) is displayed on the HUD or instrument panel to provide basic information prompts. When the risk index is greater than 0.6 and less than 0.8, the risk level can be determined as medium risk, and in addition to visual warnings, the system's recommended obstacle avoidance guide line (such as green) is highlighted, supplemented by voice prompts (such as "We suggest changing lanes later"). When the risk index is greater than 0.8, the risk level can be determined as high risk, and a strong warning is activated, including visual (flashing red guide lines or highlighted danger areas), tactile (high-frequency vibration of the steering wheel or seat), and auditory (clear, forceful voice commands, such as "Stop turning!") multi-channel warnings, and may trigger the aforementioned active collaborative control functions.

[0105] The above primarily describes the solutions provided in this application from the perspective of the device's working principle. It is understood that, to achieve the aforementioned functions, the vehicle obstacle avoidance control device includes corresponding hardware structures and / or software modules for executing each function. Those skilled in the art should readily recognize that, based on the algorithm steps of the examples described in conjunction with the embodiments disclosed herein, this application can be implemented in hardware or a combination of hardware and computer software. Whether a function is executed in hardware or by computer software driving hardware depends on the specific application and design constraints of the technical solution. Those 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 this application.

[0106] This application embodiment can divide the vehicle obstacle avoidance control device into functional modules according to the above method example. For example, each function can be divided into a separate functional module, or two or more functions can be integrated into one processing module. The integrated module can be implemented in hardware or as a software functional module.

[0107] It should be noted that the module division in this embodiment is illustrative and represents only one logical functional division; in actual implementation, other division methods may be used. When dividing functional modules according to their respective functions, Figure 5 A schematic diagram of a possible configuration of the obstacle avoidance control device for the vehicle involved in the above and embodiments is shown. Figure 5 As shown, the obstacle avoidance control device 500 of the vehicle may include: a data acquisition module 501, a prediction module 502, and a generation module 503.

[0108] The data acquisition module 501 is used to support the execution of the vehicle's obstacle avoidance control device 500. Figure 3 S301 in the illustrated obstacle avoidance control method for a vehicle.

[0109] Prediction module 502 is used to support the execution of the vehicle's obstacle avoidance control device 500. Figure 3 S302 and S303 are shown in the obstacle avoidance control method of the vehicle.

[0110] Generation module 503 is used to support the execution of the vehicle's obstacle avoidance control device 500. Figure 3 S304 in the illustrated obstacle avoidance control method for vehicles.

[0111] One possible implementation involves the prediction module, when predicting the driver's driving intention and multiple alternative driving options for the vehicle at the next moment based on vehicle and driver state information, specifically predicting the driver's driving intention based on the vehicle and driver state information. Based on the driving intention and a pre-built vehicle dynamics model, it predicts multiple alternative driving trajectories for the vehicle at the next moment, as well as the probability of driving each alternative trajectory.

[0112] One possible implementation involves the prediction module performing spatiotemporal coupling analysis between the target information and each alternative driving information to predict the risk information corresponding to each alternative driving information. Specifically, this involves predicting the target trajectory of the target object at the next moment based on the target information. For any alternative driving information among all the alternative driving information, the module performs spatiotemporal coupling analysis between the target trajectory and the alternative driving trajectory of the target information to predict the risk index of a collision between the vehicle and the target object at the next moment and the time of the collision.

[0113] In one possible implementation, the obstacle avoidance control device for vehicles provided in this application embodiment is further used to determine that there is a risk coupling point in the risk information when the risk index of the vehicle colliding with the target object at the next moment exceeds the preset risk index, and the time of the vehicle colliding with the target at the next moment is less than the safe time.

[0114] One possible implementation is that, when the generation module generates obstacle avoidance information based on vehicle status information, environmental information, and risk information, it specifically determines the vehicle guidance plan based on the environmental and risk information. Based on the vehicle status information, target information, and risk information, it determines the corresponding prompts for the guidance plan.

[0115] One possible implementation is that, when determining the vehicle guidance plan based on environmental and risk information, the generation module specifically modifies the alternative driving paths corresponding to the risk coupling points in the risk information based on the environmental information to determine the obstacle avoidance path for the vehicle in the next moment. And / or, based on the environmental and risk information, it determines operational suggestions for the driver in the next moment. And / or, when the vehicle's auxiliary control function is detected to be activated, it generates auxiliary control commands for the vehicle based on the environmental and risk information.

[0116] One possible implementation involves the generation module, when determining the prompt information corresponding to the guidance plan based on vehicle status information, target information, and risk information, specifically determining the relative speed between the vehicle and the target object based on the vehicle status information and target information. Based on the relative speed, the probability of driving the alternative driving path corresponding to the risk coupling point in the alternative driving information, and the time when the vehicle will collide with the target object in the next moment as indicated in the risk information, the vehicle's risk level is determined. Based on the risk level, the prompt information corresponding to the guidance plan is determined.

[0117] It should be noted that all relevant content of each step involved in the above method embodiments can be referenced from the functional description of the corresponding functional module, and will not be repeated here.

[0118] The obstacle avoidance control device 500 for vehicles provided in this application embodiment is used to perform the above-mentioned... Figure 3 The obstacle avoidance control method shown can therefore achieve the same effect as the obstacle avoidance control method of the vehicle described above.

[0119] This application also provides an obstacle avoidance control device for a vehicle, which can execute the obstacle avoidance control method and related steps of the vehicle in the above method embodiments.

[0120] This application also provides a computer-readable storage medium storing instructions thereon, which, when executed, perform the obstacle avoidance control method and related steps of the vehicle in the above method embodiments.

[0121] This application also provides a computer program product that, when run on a computer, causes the computer to execute the obstacle avoidance control method and related steps of the vehicle described in the above method embodiments.

[0122] In some embodiments, the methods shown in this application can be implemented as computer program instructions encoded in a machine-readable format on a computer-readable storage medium or on other non-transitory media or articles of art.

[0123] This application also provides a vehicle obstacle avoidance control system 100, such as... Figure 6As shown, the obstacle avoidance control system 100 of the vehicle includes at least one processor 601 and at least one interface circuit 602.

[0124] As an example, when the obstacle avoidance control system 100 of the vehicle includes a processor and an interface circuit, then the processor can be... Figure 6 The processor 601 shown in the solid box (or the processor 601 shown in the dashed box) can be an interface circuit. Figure 6 The interface circuit 602 is shown in the solid box (or the dashed box). When the vehicle's obstacle avoidance control system 100 includes two processors and two interface circuits, then the two processors include... Figure 6 The processor 601 shown in the solid box and the processor 601 shown in the dashed box, these two interface circuits include Figure 6 Interface circuit 602 is shown in both solid and dashed boxes. No limitations are imposed on this.

[0125] Processor 601 and interface circuit 602 can be interconnected via a line. For example, interface circuit 602 can be used to receive signals. Alternatively, interface circuit 602 can be used to send signals to other devices (e.g., processor 601). For instance, interface circuit 602 can read computer instructions stored in memory and send those instructions to processor 601. Processor 601 executes the instructions and, in conjunction with input / output devices, implements the various steps in the above embodiments, such as implementing... Figures 3-5 The methods illustrated are the steps performed in the embodiments shown. Of course, the obstacle avoidance control system of this vehicle may also include other discrete components, and this application embodiment does not specifically limit this.

[0126] Through the above description of the embodiments, those skilled in the art can clearly understand that, for the sake of convenience and brevity, only the division of the above functional modules is used as an example. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above.

[0127] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another device, or some features may be ignored or not executed. Furthermore, the mutual coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0128] The units described as separate components may or may not be physically separate. A component shown as a unit can be one or more physical units; that is, it can be located in one place or distributed in multiple different locations. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0129] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0130] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a readable storage medium. Based on this understanding, the technical solution of the embodiments of this application, or the part that contributes to it, or all or part of the technical solution, can be embodied in the form of a software product. This software product is stored in a storage medium and includes several instructions to cause a device (which may be a microcontroller, chip, etc.) or processor to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0131] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for obstacle avoidance control of a vehicle, characterized in that, The method includes: The system collects vehicle status information, driver status information, and environmental information at the current moment; the environmental information includes target information of target objects located in the vehicle's blind spot. Based on the vehicle status information and the driver status information, predict the driver's driving intention at the next moment and multiple alternative driving information for the vehicle at the next moment; The target information is spatiotemporally coupled with each of the candidate driving information to predict the risk information corresponding to each of the candidate driving information; If a risk coupling point is detected in the risk information, obstacle avoidance information is generated based on the vehicle status information, the environmental information, and the risk information to avoid the target object.

2. The method according to claim 1, characterized in that, The step of predicting the driver's driving intention at the next moment and multiple alternative driving information for the vehicle at the next moment based on the vehicle status information and the driver status information includes: Based on the vehicle status information and the driver status information, predict the driver's driving intention at the next moment; Based on the driving intention and the pre-built vehicle dynamics model, predict multiple alternative driving trajectories for the vehicle at the next moment and the driving probability of each alternative driving trajectory.

3. The method according to claim 1, characterized in that, The step of performing spatiotemporal coupling analysis between the target information and each candidate driving information to predict the risk information corresponding to each candidate driving information includes: Based on the target information, predict the target object's trajectory at the next moment; For any target alternative driving information among all alternative driving information, the target motion trajectory and the target alternative driving trajectory of the target alternative driving information are subjected to spatiotemporal coupling analysis to predict the risk index of the vehicle colliding with the target object at the next moment and the time of the collision between the vehicle and the target at the next moment.

4. The method according to claim 1, characterized in that, The method further includes: If the risk index of the vehicle colliding with the target object at the next moment exceeds the preset risk index, and the time of the collision between the vehicle and the target at the next moment is less than the safe time, it is determined that the risk coupling point exists in the risk information.

5. The method according to claim 1, characterized in that, The step of generating obstacle avoidance information based on the vehicle status information, the environmental information, and the risk information includes: Based on the environmental information and the risk information, a guidance plan for the vehicle is determined; Based on the vehicle status information, the target information, and the risk information, the prompt information corresponding to the guidance scheme is determined.

6. The method according to claim 5, characterized in that, Determining the vehicle guidance plan based on the environmental information and the risk information includes: Based on the environmental information, the alternative driving paths corresponding to the risk coupling points in the risk information are corrected to determine the obstacle avoidance path of the vehicle in the next moment; And / or, based on the environmental information and the risk information, determine the driving recommendations for the driver at the next moment; And / or, if the vehicle's auxiliary control function is detected to be activated, auxiliary control commands for the vehicle are generated based on the environmental information and the risk information.

7. The method according to claim 5, characterized in that, The step of determining the prompt information corresponding to the guidance scheme based on the vehicle status information, the target information, and the risk information includes: Based on the vehicle status information and the target information, the relative speed between the vehicle and the target object is determined; The risk level of the vehicle is determined based on the relative speed, the probability of driving the alternative driving path corresponding to the risk coupling point in the alternative driving information, and the time when the vehicle collides with the target object in the next moment in the risk information. Based on the risk level, determine the corresponding prompt information for the guidance scheme.

8. A vehicle obstacle avoidance control device, characterized in that, The device includes: The data acquisition module is used to collect vehicle status information, driver status information, and environmental information at the current moment; the environmental information includes target information of target objects located in the vehicle's blind spot. The prediction module is used to predict the driver's driving intention at the next moment and multiple alternative driving information of the vehicle at the next moment based on the vehicle status information and the driver status information. The prediction module is also used to perform spatiotemporal coupling analysis between the target information and each of the candidate driving information to predict the risk information corresponding to each of the candidate driving information. The generation module is used to generate obstacle avoidance information based on the vehicle status information, the environmental information, and the risk information when a risk coupling point is detected in the risk information, so as to avoid the target object.

9. An obstacle avoidance control device for a vehicle, characterized in that, The obstacle avoidance control device of the vehicle includes a processor and a memory, the memory storing machine-executable instructions that can be executed by the processor, and the processor executing the machine-executable instructions to implement the obstacle avoidance control method of the vehicle according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the obstacle avoidance control method for the vehicle as described in any one of claims 1 to 7.

Images