A collision mitigation system multi-execution unit fusion control and execution method and system

Abstract

The present invention provides a collision mitigation system multi-execution unit fusion control and execution method and system, belonging to the field of vehicle braking control technology, the method includes: step 1: based on the target detection system of the collision mitigation system multi-sensor fusion, identify the target collision time of the vehicle and the obstacle; step 2: obtain the current working condition; step 3: based on the safe distance model algorithm, determine the braking deceleration according to the target collision time and the current working condition; step 4: perform multi-execution unit fusion control and execution according to the braking deceleration. A collision mitigation system multi-execution unit fusion control and execution method and system of the present invention introduces a safe distance model algorithm, and determines the braking deceleration for the vehicle and the obstacle to maintain a safe distance according to the target collision time and the current working condition; according to the braking deceleration, multiple braking execution units are coordinated and executed, which improves the vehicle control stability and further improves driving safety.

Description

Technical Field

[0001] The present invention relates to the technical field of vehicle brake control, and in particular to a collision mitigation system multi-execution unit fusion control and execution method and system. Background Art

[0002] The collision mitigation system is an advanced vehicle safety technology that aims to automatically detect and predict potential collision risks through sensors and intelligent systems, and take proactive measures to reduce the probability and severity of collisions. Currently, most commercial and passenger car collision mitigation systems use a single brake actuator unit (EBS or valve or motor) to implement automatic braking. EBS is an advanced electronically controlled braking system that controls the brake pressure through electronic signals, thereby improving the response speed and accuracy of braking. The pneumatic valve controls the size of the braking force by adjusting the air pressure, while the motor directly drives the brake to achieve a fast braking response.

[0003] However, these actuators may lack braking force after running for a long time, or even malfunction and fail to respond to automatic braking commands, increasing the risk of traffic accidents.

[0004] In view of this, there is an urgent need for a collision mitigation system multi-execution unit fusion control and execution method and system to at least solve the above-mentioned deficiencies. Summary of the invention

[0005] One of the purposes of the present invention is to provide a collision mitigation system multi-execution unit fusion control and execution method and system, which identifies the target collision time based on the target detection system of multi-sensor fusion; introduces a safe distance model algorithm, and determines the braking deceleration for maintaining a safe distance between the vehicle and the obstacle based on the target collision time and the current working conditions; and based on the braking deceleration, performs coordinated control and execution of multiple braking execution units, thereby avoiding the situation where a single brake actuator has insufficient braking force or fails to respond to the braking command, thereby improving the vehicle control stability and further improving driving safety.

[0006] An embodiment of the present invention provides a collision mitigation system multi-execution unit fusion control and execution method, including:

[0007] Step 1: The target detection system based on the multi-sensor fusion of the collision mitigation system identifies the target collision time between the vehicle and the obstacle;

[0008] Step 2: Get the current working conditions;

[0009] Step 3: Based on the safety distance model algorithm, determine the braking deceleration according to the target collision time and current working conditions;

[0010] Step 4: Perform multi-execution unit fusion control and execution according to the braking deceleration.

[0011] Preferably, step 1: identifying the target collision time between the vehicle and the obstacle based on the target detection system of the collision mitigation system multi-sensor fusion, comprises:

[0012] Acquire visual detection data from visual sensors;

[0013] Obtain radar detection data of millimeter wave radar;

[0014] Based on the visual detection data and radar detection data, the target collision time is identified.

[0015] Preferably, step 4: performing fusion control and execution of multiple execution units according to the braking deceleration includes:

[0016] Output braking force according to braking deceleration;

[0017] Monitor actual braking force and adjust it dynamically;

[0018] Determining the fusion braking force of multiple execution units according to the dynamic adjustment result of the actual braking force dynamic adjustment;

[0019] According to the fusion braking force, the brake components corresponding to the execution unit are controlled accordingly.

[0020] Preferably, outputting the braking force according to the braking deceleration includes:

[0021] Determine the required braking force according to the braking deceleration;

[0022] Characterizing the required braking force based on a preset first characterization rule to obtain required braking force characteristics, where the required braking force characteristics include: required braking force direction and required braking force value;

[0023] According to the required braking force characteristics, a target braking force representation vector is constructed;

[0024] Obtain the braking parameters of the execution unit;

[0025] Characterizing the braking parameters according to the first characterization rule to obtain a braking force characteristic, wherein the braking force characteristic includes: a braking force direction range and a braking force value range of the execution unit;

[0026] According to the braking force characteristics, construct a representation vector of the braking force to be synthesized;

[0027] Determining a synthesis scheme of the braking force representation vector to be synthesized of the execution unit based on the target braking force representation vector and the braking force representation vector to be synthesized;

[0028] Based on the synthesis scheme, calculate the braking energy consumption;

[0029] Screening the target synthesis scheme with minimum braking energy consumption;

[0030] The braking force is output according to the target synthesis scheme.

[0031] Preferably, determining a synthesis scheme of the braking force representation vector to be synthesized of the execution unit based on the target braking force representation vector and according to the braking force representation vector to be synthesized includes:

[0032] Obtain a three-dimensional model of the vehicle;

[0033] The target braking force representation vector and the braking force representation vector to be synthesized are marked correspondingly in the vehicle three-dimensional model to obtain a marking diagram;

[0034] Characterizing the labeled graph based on a preset second characterization rule to obtain a labeled feature;

[0035] Machine learning force analysis rules to obtain force analysis models;

[0036] Input the marked features into the force analysis model to obtain the force analysis diagram;

[0037] Extract the synthesis scheme according to the force analysis diagram and the preset synthesis scheme extraction rules.

[0038] Preferably, monitoring the actual braking force and performing dynamic adjustment includes:

[0039] According to the monitoring result of the actual braking force dynamic adjustment, it is determined whether there is a brake execution unit failure;

[0040] If yes, report the fault information of the corresponding brake execution unit to the cloud platform and recalculate the braking force;

[0041] Distribute the recalculated braking force to the new brake actuation unit;

[0042] The new brake actuating unit is controlled to perform a braking operation according to the allocated recalculated braking force.

[0043] Preferably, step 3: determining the braking deceleration based on the safety distance model algorithm according to the target collision time and the current working condition includes:

[0044] Based on the safety distance model algorithm, the pre-selected braking deceleration is determined according to the target collision time and the current working conditions;

[0045] If the preselected braking deceleration is greater than the preset braking deceleration threshold, an avoidance path is planned according to the current working conditions.

[0046] Preferably, planning an avoidance path according to the current working conditions includes:

[0047] Get the adjacent lane information of the vehicle's current lane;

[0048] Characterizing the adjacent lane information based on a preset third characterization rule to obtain adjacent lane information features, the adjacent lane information features including: adjacent lane image features and adjacent lane sign recognition results;

[0049] Obtain lane change condition information;

[0050] Characterizing the lane changing condition information based on a third characterization rule to obtain lane changing condition information features;

[0051] Performing information feature matching on the lane change condition information feature and the adjacent lane information feature; wherein the source image information acquisition device of the lane change condition information feature and the adjacent lane information feature for information feature matching is consistent;

[0052] If there is an information feature matching value greater than or equal to the information feature matching value threshold preset by the corresponding source image information acquisition device, obtaining the information feature matching result of the auxiliary image information acquisition device of the corresponding source image information acquisition device;

[0053] If there is no preset lane change conflict result in the information feature matching result of the auxiliary image information acquisition device, it is determined that the adjacent lane meets the lane change condition, and the vehicle is controlled to change lanes to the adjacent lane that meets the lane change condition.

[0054] Preferably, obtaining lane change condition information includes:

[0055] Acquire device parameters of an image information acquisition device in adjacent lane information;

[0056] determining a target position of the image information acquisition device relative to the vehicle;

[0057] Get safe lane change distance;

[0058] Determine the detection distance of the detection target from the vehicle in the image captured by the device according to the device parameters and the target position;

[0059] The images captured by the device whose detection distance is greater than or equal to the safe lane-changing distance are screened as image information in the lane-changing condition information.

[0060] An embodiment of the present invention provides a collision mitigation system multi-execution unit fusion control and execution system, including:

[0061] The target collision time identification subsystem is used for the target detection system based on the multi-sensor fusion of the collision mitigation system to identify the target collision time between the vehicle and the obstacle;

[0062] The current working condition acquisition subsystem is used to obtain the current working condition;

[0063] The braking deceleration determination subsystem is used to determine the braking deceleration based on the safety distance model algorithm according to the target collision time and the current working condition;

[0064] The execution subsystem is used to perform fusion control and execution of multiple execution units according to the braking deceleration.

[0065] The beneficial effects of the present invention are:

[0066] The present invention identifies the target collision time according to the target detection system of multi-sensor fusion; introduces a safe distance model algorithm, and determines the braking deceleration for maintaining a safe distance between the vehicle and the obstacle according to the target collision time and the current working conditions; and performs coordinated control and execution of multiple braking execution units according to the braking deceleration, thereby avoiding the situation where a single braking execution mechanism has insufficient braking force or fails to respond to the braking command, thereby improving the vehicle control stability and further improving driving safety.

[0067] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures specifically pointed out in the present application documents.

[0068] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0070] Figure 1 A schematic diagram of a multi-execution unit fusion control and execution method of a collision mitigation system in an embodiment of the present invention;

[0071] Figure 2 Schematic diagram of a multi-execution unit fusion control and execution system of a collision mitigation system in an embodiment of the present invention. DETAILED DESCRIPTION

[0072] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0073] The embodiment of the present invention provides a collision mitigation system multi-execution unit fusion control and execution method, such as Figure 1 As shown, including:

[0074] Step 1: Based on the target detection system of the collision mitigation system multi-sensor fusion, the target collision time between the vehicle and the obstacle is identified; wherein the collision mitigation system is: the vehicle's active safety system, including sensors, controllers and actuators, which can automatically or assist the driver in braking and steering operations; the target detection system is: a system for detecting obstacles on the way of the vehicle, based on multi-sensor fusion data (such as: millimeter wave radar, images during driving); the target collision time is: the estimated time when the obstacle detected by the target detection system collides with the vehicle, which is calculated by measuring the distance, speed and relative motion state of the obstacle;

[0075] Step 2: Obtain current operating conditions; wherein the current operating conditions are: the current operating state and environmental conditions of the vehicle, including but not limited to the vehicle's speed, acceleration, steering angle, road conditions, and weather conditions;

[0076] Step 3: Based on the safety distance model algorithm, determine the braking deceleration according to the target collision time and the current working condition; the safety distance model algorithm is: an AI model algorithm that calculates the braking deceleration applied when the vehicle and the obstacle maintain a safe distance (for example, 0.5 meters) based on the target collision time and the current working condition, and obtains the model training based on multiple historical working conditions and the deceleration records corresponding to the historical working conditions that meet the manually preset safety distance standards;

[0077] Step 4: Based on the braking deceleration, perform multi-execution unit fusion control and execution. Among them, the multi-execution unit is: multiple braking execution units (such as: EBS or valve or motor). The multi-execution unit fusion control and execution based on the braking deceleration means: when a single braking execution unit fails and the collision mitigation system cannot brake, the collision mitigation system continues to take effect and the vehicle is controlled smoothly by switching the execution unit and adjusting the control algorithm.

[0078] The working principle and beneficial effects of the above technical solution are:

[0079] The present invention identifies the target collision time according to the target detection system of multi-sensor fusion; introduces a safe distance model algorithm, and determines the braking deceleration for maintaining a safe distance between the vehicle and the obstacle according to the target collision time and the current working conditions; and performs coordinated control and execution of multiple braking execution units according to the braking deceleration, thereby avoiding the situation where a single braking execution mechanism has insufficient braking force or fails to respond to the braking command, thereby improving the vehicle control stability and further improving driving safety.

[0080] In one embodiment, step 1: identifying a target collision time between a vehicle and an obstacle based on a target detection system of a collision mitigation system multi-sensor fusion, including:

[0081] Obtain visual detection data from a visual sensor; wherein the visual sensor is: a camera; the visual detection data is: the shape, color, position and motion state of the detection object obtained after processing and analyzing the image data obtained by the visual sensor;

[0082] Acquire radar detection data of a millimeter-wave radar; wherein the radar detection data is: scanning data of the millimeter-wave radar;

[0083] Based on the visual detection data and radar detection data, the target collision time is identified.

[0084] The working principle and beneficial effects of the above technical solution are:

[0085] The present invention combines the visual detection data of the visual sensor and the radar detection data of the millimeter wave radar to perform target detection and calculate the target collision time. The target collision time is calculated more accurately, further improving the calculation accuracy of the subsequent braking deceleration.

[0086] In one embodiment, step 4: performing multi-execution unit fusion control and execution according to the braking deceleration includes:

[0087] Outputting the braking force according to the braking deceleration; wherein the outputting the braking force is: firstly determining the braking force output schemes of the plurality of execution units according to the braking deceleration, and then outputting the braking force according to the braking force output scheme;

[0088] Monitor the actual braking force and make dynamic adjustments; the actual braking force is the braking force actually applied to the vehicle by the execution unit after the braking force is output according to the braking force output plan; dynamic adjustment is the process of excluding the braking force output of the faulty node and automatically allocating the braking force originally allocated to the faulty node to other execution units when insufficient braking force or failure of the execution unit is detected;

[0089] According to the dynamic adjustment result of the actual braking force dynamic adjustment, the fused braking force of multiple execution units is determined; wherein the dynamic adjustment result is: a reallocation scheme of the braking force originally allocated to the faulty node to other execution units; the fused braking force of multiple execution units is determined as: the final execution unit and the braking sub-strategy of the execution unit determined by combining the initial braking force allocation scheme of the non-faulty node and the reallocation scheme of the faulty node;

[0090] According to the fusion braking force, the brake components corresponding to the execution unit are controlled accordingly.

[0091] The working principle and beneficial effects of the above technical solution are:

[0092] The present invention dynamically adjusts the actual braking force by monitoring the actual deceleration feedback of the vehicle, thereby avoiding the situation where a single brake actuator may have insufficient braking force and cannot respond to automatic braking when a fault occurs, thereby improving driving safety.

[0093] In one embodiment, outputting the braking force according to the braking deceleration includes:

[0094] Determine the required braking force according to the braking deceleration; wherein the required braking force is calculated and determined according to the braking deceleration and the vehicle mass;

[0095] Characterizing the required braking force based on a preset first characterization rule to obtain the required braking force characteristics, the required braking force characteristics including: a required braking force direction and a required braking force value; wherein the preset first characterization rule is: a rule for obtaining the magnitude and direction of the required braking force; the required braking force direction is, for example, opposite to the rotation direction of the wheel; the required braking force value is, for example, 3000N;

[0096] According to the required braking force characteristics, a target braking force representation vector is constructed; wherein the direction of the target braking force representation vector represents the direction of action of the required braking force, and the modulus length of the target braking force representation vector represents the numerical value of the required braking force;

[0097] Obtaining the braking parameters of the execution unit; wherein the braking parameters are: the direction range and value range of the braking force exerted by the execution unit on the vehicle;

[0098] Characterizing the braking parameters according to the first characterization rule to obtain a braking force characteristic, wherein the braking force characteristic includes: a braking force direction range and a braking force value range of the execution unit;

[0099] According to the braking force characteristics, a characterization vector of the braking force to be synthesized is constructed; wherein the characterization vector of the braking force to be synthesized is a vector of the direction of action of the braking force that can be generated by the direction characterization execution unit and the numerical value of the braking force that can be generated by the modulus length characterization execution unit;

[0100] Based on the target braking force representation vector and the braking force representation vector to be synthesized, a synthesis scheme of the braking force representation vector to be synthesized of the execution unit is determined; wherein the synthesis scheme is: a scheme for synthesizing the target braking force representation vector according to the braking force representation vector to be synthesized, including multiple target braking force representation vectors to be synthesized;

[0101] Based on the synthesis scheme, the braking energy consumption is calculated; wherein the braking energy consumption is based on the target braking force characterization vector to be synthesized, each target braking force characterization vector to be synthesized corresponds to a preset braking energy consumption value, and the braking energy consumption is obtained by accumulating the braking energy consumption values ​​corresponding to the target braking force characterization vector to be synthesized;

[0102] Screening the target synthesis scheme with minimum braking energy consumption;

[0103] The braking force is output according to the target synthesis scheme.

[0104] The working principle and beneficial effects of the above technical solution are:

[0105] The present invention characterizes the required braking force calculated according to the braking deceleration to obtain the required braking force characteristics, and constructs a target braking force characterization vector describing the required braking force; characterizes the braking parameters of the execution unit to obtain the braking force characteristics, and constructs a braking force characterization vector to be synthesized that describes the braking force that the execution unit can apply, and each execution unit corresponds to a plurality of braking force characterization vectors to be synthesized; synthesizes the target braking force characterization vector according to the braking force characterization vector to be synthesized, and determines a synthesis scheme, which includes a plurality of target braking force characterization vectors to be synthesized, and each target braking force characterization vector to be synthesized corresponds to an execution unit; however, there may be multiple synthesis schemes, and therefore, it is necessary to screen a suitable synthesis scheme, calculate the braking energy consumption corresponding to the synthesis scheme, and output the braking force according to the target synthesis scheme, thereby improving the rationality of the braking force distribution.

[0106] In one embodiment, determining a synthesis scheme of the braking force representation vector to be synthesized of the execution unit based on the target braking force representation vector and according to the braking force representation vector to be synthesized includes:

[0107] Obtain a three-dimensional model diagram of a vehicle; wherein the three-dimensional model diagram of the vehicle is: a three-dimensional stereoscopic model diagram of the vehicle;

[0108] The target braking force representation vector and the braking force representation vector to be synthesized are marked correspondingly in the three-dimensional vehicle model to obtain a marked diagram; wherein the marked point position (vector starting point) of the target braking force representation vector is at the center of gravity of the vehicle in the three-dimensional vehicle model diagram; the marked point position of the braking force representation vector to be synthesized in the three-dimensional vehicle model diagram is determined according to the braking force action point of the execution unit corresponding to the braking force representation vector to be synthesized on the vehicle;

[0109] Characterizing the annotated graph based on a preset second characterization rule to obtain annotated features; wherein the second characterization rule is a rule for extracting information related to force analysis in the annotated graph, and the annotated features are: the annotated position relationship between the target braking force representation vector, the braking force representation vector to be synthesized, and the vehicle three-dimensional model graph;

[0110] The machine learns stress analysis rules to obtain a stress analysis model; wherein the stress analysis model is a deep learning model that uses machine learning stress analysis rules to replace manual stress analysis;

[0111] The annotated features are input into the force analysis model to obtain a force analysis diagram; wherein the force analysis diagram is the result of determining a braking force representation vector to be synthesized (target synthetic braking force representation vector) that can achieve the target by taking the target braking force representation vector as the target for the final synthesis of the braking force representation vector to be synthesized and annotating it in the vehicle three-dimensional model diagram;

[0112] The synthesis scheme is extracted according to the force analysis diagram and the preset synthesis scheme extraction rule, wherein the preset synthesis scheme extraction rule is: a rule for reading the execution logic of each execution unit according to the target synthetic braking force representation vector.

[0113] The working principle and beneficial effects of the above technical solution are:

[0114] The present invention marks the target braking force characterization vector and the braking force characterization vector to be synthesized correspondingly in the vehicle three-dimensional model to obtain a marked graph, and introduces a second characterization rule to characterize the marked graph to obtain a marked feature; the machine learns the force analysis rule to obtain a force analysis model, and the force analysis model replaces manual work to determine the object's force scheme (force analysis graph) according to the object's force requirements (target braking force characterization vector, extracted from the marked features) and force conditions (braking force characterization vector to be synthesized, extracted from the marked features), and extracts the synthesis scheme according to the force analysis graph, which is more intelligent.

[0115] In one embodiment, monitoring the actual braking force and making dynamic adjustments include:

[0116] According to the monitoring result of the actual braking force dynamic adjustment, it is judged whether there is a brake execution unit failure; wherein the monitoring result is: whether the braking force actually output by the execution unit is consistent with the braking force allocated to it;

[0117] If yes, report the fault information of the corresponding braking execution unit to the cloud platform and recalculate the braking force; wherein recalculating the braking force is: after detecting the braking force output of the faulty node, automatically reallocate the braking force originally allocated to the faulty node to other execution units;

[0118] Distribute the recalculated braking force to the new brake actuation unit;

[0119] The new brake actuating unit is controlled to perform a braking operation according to the allocated recalculated braking force.

[0120] The working principle and beneficial effects of the above technical solution are:

[0121] The present invention monitors the operating status of each execution unit in real time. If a certain execution unit fails, the fault will be reported to the cloud platform, and the control algorithm will be adjusted to eliminate the braking force output of the faulty node. The braking force originally allocated to the faulty node will be automatically allocated to other units, thereby automatically switching the execution unit in the event of a failure, improving the stability of vehicle control and further improving driving safety.

[0122] In one embodiment, step 3: determining the braking deceleration based on the safety distance model algorithm according to the target collision time and the current working condition includes:

[0123] Based on the safety distance model algorithm, the preselected braking deceleration is determined according to the target collision time and the current working condition; wherein the preselected braking deceleration is: the braking deceleration required for safe braking of the vehicle output by the safety distance model algorithm according to the target collision time and the current working condition;

[0124] If the preselected braking deceleration is greater than the preset braking deceleration threshold, an avoidance path is planned according to the current working condition; wherein the preset braking deceleration threshold is the maximum braking deceleration that the vehicle can apply;

[0125] Among them, planning the avoidance path according to the current working conditions includes:

[0126] Obtain the adjacent lane information of the vehicle's current lane; the adjacent lane information includes: lane identification information (e.g., dashed lane line) and lane image information (e.g., adjacent lane image captured by a 360-degree panoramic camera or a streaming rearview mirror) of the adjacent lane of the vehicle's current lane.

[0127] Characterizing the adjacent lane information based on a preset third characterization rule to obtain adjacent lane information features, the third characterization rule being: a rule for extracting image features of the adjacent lane information and matching the image features of the adjacent lane information with pre-marked adjacent lane identification image features; the adjacent lane information features include: adjacent lane image features and adjacent lane identification recognition results;

[0128] Acquire lane change condition information; wherein the lane change condition information includes: lane image information of lanes that allow lane change and adjacent lane identification information, such as: lane image of an adjacent lane with a dashed lane line close to the current lane and no vehicle behind;

[0129] Characterizing the lane change condition information based on the third characterization rule to obtain lane change condition information features; wherein the lane change condition information features are: a characteristic representation of the lane change condition information, such as: relevant information indicating that a lane line of an adjacent lane close to the current lane is a dotted line, and lane image information indicating that there is no vehicle behind;

[0130] Performing information feature matching on the lane change condition information feature and the adjacent lane information feature; wherein the source image information acquisition device of the lane change condition information feature and the adjacent lane information feature for information feature matching is consistent;

[0131] If there is an information feature matching value greater than or equal to the preset information feature matching value threshold of the corresponding source image information acquisition device, the information feature matching result of the auxiliary image information acquisition device of the corresponding source image information acquisition device is obtained; wherein the preset information feature matching value threshold is manually preset; the auxiliary image information acquisition device of the source image information acquisition device is: an image information acquisition device that assists in shooting the surrounding environment (blind spot) of the vehicle body that the source image information acquisition device cannot shoot, for example: the source image information acquisition device is a camera under the rearview mirror, and the auxiliary shooting source image information acquisition device is a camera set on the B-pillar of the vehicle body;

[0132] If there is no preset lane change conflict result in the information feature matching result of the auxiliary image information acquisition device, it is determined that the adjacent lane meets the lane change condition, and the vehicle is controlled to change lanes to the adjacent lane that meets the lane change condition; the lane change conflict result is, for example: there are obstacles in the surrounding environment of the vehicle body that cannot be captured by the auxiliary shooting source image information acquisition device;

[0133] Among them, obtaining lane change condition information includes:

[0134] Obtaining device parameters of an image information acquisition device in the adjacent lane information; wherein the image information acquisition device is: an acquisition device of image information in the adjacent lane information, such as: a wide-angle camera under a rearview mirror, a streaming media rearview mirror, and an electronic rearview mirror; the device parameters are: parameters of the image information acquisition device, such as: focal length;

[0135] Determine a target position of the image information acquisition device relative to the vehicle; wherein the target position is: the position of the image information acquisition device relative to the vehicle, such as: below the rearview mirror, on the B-pillar of the vehicle;

[0136] Obtain a safe lane change distance; wherein the safe lane change distance is preset manually, for example: 15 meters;

[0137] Determine the detection distance of the detection target in the image captured by the device from the vehicle according to the device parameters and the target position; wherein the detection distance is: the distance between the detection target and the rear end of the vehicle. When determining the detection distance, first calculate the first distance between the detection target and the target position using a deep learning model according to the device parameters and the image captured by the device, then determine the second distance from the rear end of the vehicle to the target position according to the vehicle body parameters and the target position, and obtain the detection distance by subtracting the second distance from the first distance;

[0138] The images captured by the device whose detection distance is greater than or equal to the safe lane-changing distance are screened as image information in the lane-changing condition information.

[0139] The working principle and beneficial effects of the above technical solution are:

[0140] Although multiple execution units improve the flexibility of vehicle braking, the braking capacity of the vehicle has an upper limit. When the pre-selected braking deceleration calculated by the safety distance model algorithm based on the target collision time and the current working condition is greater than the braking deceleration threshold, the vehicle cannot successfully brake before the collision occurs. Therefore, the adjacent lane information of the vehicle's current lane (lane identification information and lane image information of the adjacent lane) is obtained, and the third characterization rule is introduced to characterize the adjacent lane information to obtain the adjacent lane information features;

[0141] Acquire lane change condition information; when acquiring the lane change condition information, determine the detection distance of the detection target from the vehicle in the image captured by the device according to the different device parameters of the image information acquisition device in the lane change condition information and the different positions relative to the vehicle body, and select the images captured by the device with the detection distance greater than or equal to the safe lane change distance for subsequent extraction of image features for allowing lane change, thereby improving the rationality of the basis for lane change conditions;

[0142] The lane changing condition information is also characterized to obtain the lane changing condition information features; the lane changing condition information features are matched with the adjacent lane information features. When the adjacent lane meets the lane changing conditions, the vehicle is controlled to change lanes to the adjacent lane that meets the lane changing conditions. When the braking capacity is insufficient, reasonable lane changes are made to improve safety.

[0143] The embodiment of the present invention provides a collision mitigation system multi-execution unit fusion control and execution system, such as Figure 2 As shown, including:

[0144] A target collision time identification subsystem 1, which is used for identifying the target collision time between the vehicle and the obstacle based on the target detection system of the collision mitigation system multi-sensor fusion;

[0145] Current working condition acquisition subsystem 2, used to acquire the current working condition;

[0146] A braking deceleration determination subsystem 3, for determining the braking deceleration based on a safety distance model algorithm according to the target collision time and the current working condition;

[0147] The execution subsystem 4 is used to perform fusion control and execution of multiple execution units according to the braking deceleration.

[0148] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A collision mitigation system multi-execution unit fusion control and execution method, characterized in that: include: Step 1: The target detection system based on the multi-sensor fusion of the collision mitigation system identifies the target collision time between the vehicle and the obstacle; Step 2: Get the current working conditions; Step 3: Based on the safety distance model algorithm, determine the braking deceleration according to the target collision time and current working conditions; Step 4: Determine the required braking force based on the braking deceleration; Characterizing the required braking force based on a preset first characterization rule to obtain required braking force characteristics, where the required braking force characteristics include: required braking force direction and required braking force value; According to the required braking force characteristics, a target braking force representation vector is constructed; Obtain the braking parameters of the execution unit; Characterizing the braking parameters according to the first characterization rule to obtain a braking force characteristic, wherein the braking force characteristic includes: a braking force direction range and a braking force value range of the execution unit; According to the braking force characteristics, construct a representation vector of the braking force to be synthesized; Obtain a three-dimensional model of the vehicle; The target braking force representation vector and the braking force representation vector to be synthesized are marked correspondingly in the vehicle three-dimensional model to obtain a marking diagram; Characterizing the labeled graph based on a preset second characterization rule to obtain a labeled feature; Machine learning force analysis rules to obtain force analysis models; Input the marked features into the force analysis model to obtain the force analysis diagram; Extract the synthesis scheme according to the force analysis diagram and the preset synthesis scheme extraction rules; Based on the synthesis scheme, calculate the braking energy consumption; Screening the target synthesis scheme with minimum braking energy consumption; Output braking force according to a target synthesis scheme; Monitor actual braking force and adjust it dynamically; Determining the fusion braking force of multiple execution units according to the dynamic adjustment result of the actual braking force dynamic adjustment; According to the fusion braking force, the brake components corresponding to the execution unit are controlled accordingly.

2. A collision mitigation system multi-execution unit fusion control and execution method as claimed in claim 1, characterized in that: Step 1: The target detection system based on the multi-sensor fusion of the collision mitigation system identifies the target collision time between the vehicle and the obstacle, including: Acquire visual detection data from visual sensors; Obtain radar detection data of millimeter wave radar; Based on the visual detection data and radar detection data, the target collision time is identified.

3. The collision mitigation system multi-execution unit fusion control and execution method according to claim 1, characterized in that: Monitors actual braking force and adjusts it dynamically, including: According to the monitoring result of the actual braking force dynamic adjustment, it is determined whether there is a brake execution unit failure; If yes, report the fault information of the corresponding brake execution unit to the cloud platform and recalculate the braking force; Distribute the recalculated braking force to the new brake actuation unit; The new brake actuating unit is controlled to perform a braking operation according to the allocated recalculated braking force.

4. The collision mitigation system multi-execution unit fusion control and execution method according to claim 1, characterized in that: Step 3: Based on the safety distance model algorithm, determine the braking deceleration according to the target collision time and current working conditions, including: Based on the safety distance model algorithm, the pre-selected braking deceleration is determined according to the target collision time and the current working conditions; If the preselected braking deceleration is greater than the preset braking deceleration threshold, an avoidance path is planned according to the current working conditions.

5. A collision mitigation system multi-execution unit fusion control and execution method as claimed in claim 4, characterized in that: Plan an avoidance path based on the current working conditions, including: Get the adjacent lane information of the vehicle's current lane; Characterizing the adjacent lane information based on a preset third characterization rule to obtain adjacent lane information features, the adjacent lane information features including: adjacent lane image features and adjacent lane sign recognition results; Obtain lane change condition information; Characterizing the lane changing condition information based on a third characterization rule to obtain lane changing condition information features; Performing information feature matching on the lane change condition information feature and the adjacent lane information feature; wherein the source image information acquisition device of the lane change condition information feature and the adjacent lane information feature for information feature matching is consistent; If there is an information feature matching value greater than or equal to the information feature matching value threshold preset by the corresponding source image information acquisition device, obtaining the information feature matching result of the auxiliary image information acquisition device of the corresponding source image information acquisition device; If there is no preset lane change conflict result in the information feature matching result of the auxiliary image information acquisition device, it is determined that the adjacent lane meets the lane change condition, and the vehicle is controlled to change lanes to the adjacent lane that meets the lane change condition.

6. A collision mitigation system multi-execution unit fusion control and execution method as claimed in claim 5, characterized in that: Obtain lane change condition information, including: Obtain device parameters of an image information acquisition device in adjacent lane information; determining a target position of the image information acquisition device relative to the vehicle; Get safe lane change distance; Determine the detection distance of the detection target from the vehicle in the image captured by the device according to the device parameters and the target position; The images captured by the device whose detection distance is greater than or equal to the safe lane-changing distance are screened as image information in the lane-changing condition information.

7. A collision mitigation system multi-execution unit fusion control and execution system, characterized in that: include: The target collision time identification subsystem is used for the target detection system based on the multi-sensor fusion of the collision mitigation system to identify the target collision time between the vehicle and the obstacle; The current working condition acquisition subsystem is used to obtain the current working condition; The braking deceleration determination subsystem is used to determine the braking deceleration based on the safety distance model algorithm according to the target collision time and the current working condition; The execution subsystem is used to perform multi-execution unit fusion control and execution according to the braking deceleration, and performs the following operations: Determine the required braking force according to the braking deceleration; Characterizing the required braking force based on a preset first characterization rule to obtain required braking force characteristics, where the required braking force characteristics include: required braking force direction and required braking force value; According to the required braking force characteristics, a target braking force representation vector is constructed; Obtain the braking parameters of the execution unit; Characterizing the braking parameters according to the first characterization rule to obtain a braking force characteristic, wherein the braking force characteristic includes: a braking force direction range and a braking force value range of the execution unit; According to the braking force characteristics, construct a representation vector of the braking force to be synthesized; Obtain a three-dimensional model of the vehicle; The target braking force representation vector and the braking force representation vector to be synthesized are marked correspondingly in the vehicle three-dimensional model to obtain a marking diagram; Characterizing the labeled graph based on a preset second characterization rule to obtain a labeled feature; Machine learning force analysis rules to obtain force analysis models; Input the marked features into the force analysis model to obtain the force analysis diagram; Extract the synthesis scheme according to the force analysis diagram and the preset synthesis scheme extraction rules; Based on the synthesis scheme, calculate the braking energy consumption; Screening the target synthesis scheme with minimum braking energy consumption; Output braking force according to a target synthesis scheme; Monitor actual braking force and adjust it dynamically; Determining the fusion braking force of multiple execution units according to the dynamic adjustment result of the actual braking force dynamic adjustment; According to the fusion braking force, the brake components corresponding to the execution unit are controlled accordingly.

Images