Search

CN-122009179-A - Garage auxiliary driving method and device based on multi-domain fusion control, vehicle and medium

CN122009179ACN 122009179 ACN122009179 ACN 122009179ACN-122009179-A

Abstract

The invention relates to the technical field of vehicle control, and discloses a garage auxiliary driving method, device, vehicle and medium based on multi-domain fusion control, wherein the method comprises the steps of judging whether a driving scene of a vehicle is positioned in a garage according to multi-mode information of the vehicle; when the driving scene is in the garage, the braking system, the driving system and the suspension system of the vehicle are coordinated to prevent slipping. According to the method, the garage scene is accurately identified through multi-mode information, the linkage braking, driving and suspension systems are cooperatively controlled, and the problem of low adhesion coefficient of a wet and slippery road surface of the garage is solved. Breaks through the control limitation of a single system, inhibits the risks of skidding, tail flicking and head pushing of the vehicle from multiple dimensions, does not need manual intervention of a driver, reduces the operation burden of the driver, and greatly improves the driving safety and stability of the wet slide garage.

Inventors

  • HE BIN
  • LU TIANYI
  • LIU JIAYONG

Assignees

  • 重庆长安汽车股份有限公司

Dates

Publication Date
20260512
Application Date
20260306

Claims (10)

  1. 1. The garage assisted driving method based on multi-domain fusion control is characterized by comprising the following steps of: judging whether a driving scene of the vehicle is positioned in a garage or not according to the multi-mode information of the vehicle; When the driving scene is located in the garage, the braking system, the driving system and the suspension system of the vehicle are coordinated to prevent slipping.
  2. 2. The method according to claim 1, wherein the determining whether the driving scene of the vehicle is located in the garage according to the multi-modal information of the vehicle comprises: calculating sub-confidence scores of the driving scene belonging to the garage by utilizing each single-mode information of the vehicle respectively; Fusing the sub-confidence coefficients to obtain a total confidence score; And when the total confidence score is larger than a preset confidence threshold value, determining that the driving scene of the vehicle is positioned in the garage.
  3. 3. The method according to claim 2, wherein calculating the sub-confidence score of the driving scene belonging to the garage using the single-mode information of the vehicle, respectively, comprises: Calculating a first sub-confidence score according to collected sign visual data, wherein the sign visual data comprises a garage gate lifting rod, a height limiting mark and a speed limiting mark; calculating a second sub-confidence score through the light sensation signal intensity detected by the vehicle glass position; calculating a third sub-confidence score based on the strength of the navigational positioning signals; Calculating the confidence score of the environment where the vehicle is located as a closed environment according to the three-dimensional map data perceived by the surrounding environment of the vehicle body, and taking the confidence score as a fourth sub-confidence score; counting the number of perceived scribing parking spaces, and calculating a fifth sub-confidence score according to the size relation between the number of scribing parking spaces and a preset parking space threshold; Matching the road type of the road in the intelligent driving assistance system map by using the current vehicle position, and calculating a sixth sub-confidence score according to the road type; Calculating the confidence score of the current vehicle in the tunnel in the intelligent driving auxiliary system map by using the current vehicle position as a seventh sub-confidence score; And calculating an eighth sub-confidence score according to the vehicle speed.
  4. 4. The method of claim 1, wherein the coordinated control of the braking system, the driving system, and the suspension system of the vehicle when the driving scenario is in the garage comprises: the method comprises the steps of distributing the braking force of a vehicle and a braking execution object; controlling torque output, torque distribution and energy recovery; When the electric control suspension is fault-free and the vehicle does not trigger an emergency working condition, controlling the damping of the shock absorber to be adjusted to a preset high damping mode, and controlling the height of the suspension to be adjusted to a preset low posture mode.
  5. 5. The method according to claim 4, wherein the assigning of the magnitude of the braking force of the vehicle and the braking execution object includes: When a brake pedal depression signal is received, distributing braking forces of a front axle and a rear axle according to axle load distribution conditions of the front axle and the rear axle of the vehicle, wherein the braking force distribution ratio of the front axle to the rear axle is less than or equal to the axle load ratio of the front axle to the rear axle; when a brake pedal depression signal is received, reversely distributing mechanical braking force to each wheel according to the proportion between the real-time slip rates of each wheel; When a brake pedal depression signal is received, the braking force demand is distributed to the hydraulic or mechanical braking system and the anti-tug braking demand for the drive motor is canceled.
  6. 6. The method of claim 4, wherein said controlling torque output, torque distribution and energy recovery comprises: performing discount processing on a power torque output Map corresponding to the stroke of an accelerator pedal; when the yaw of the vehicle is not monitored, reversely distributing driving moment for each wheel according to the proportion between the real-time slip rates of each wheel; when the yaw of the vehicle is monitored, reversely distributing driving moment for each wheel according to the proportion between the real-time slip rates of each wheel, and then adjusting the driving moment of each wheel according to the yaw direction; When the travel of the accelerator pedal is detected to be larger than a preset travel threshold under the starting working condition, reducing the driving torque to establish gradient; The energy recovery function is turned off.
  7. 7. The method according to claim 1, wherein the method further comprises: when the driving scene is located in a garage, outputting first prompt information; and outputting second prompt information when the driving scene is changed from being positioned in the garage to being driven out of the garage.
  8. 8. A garage assisted driving device based on multi-domain fusion control, the device comprising: the garage identification module is used for judging whether the driving scene of the vehicle is positioned in the garage or not according to the multi-mode information of the vehicle; And the prevention control module is used for carrying out coordinated control on a braking system, a driving system and a suspension system of the vehicle when the driving scene is positioned in the garage so as to prevent slipping.
  9. 9. A vehicle, characterized by comprising: a memory and a processor in communication with each other, the memory having stored therein computer instructions which, upon execution, cause the processor to perform the method of any of claims 1 to 7.
  10. 10. A computer readable storage medium having stored thereon computer instructions for causing a computer to perform the method of any one of claims 1 to 7.

Description

Garage auxiliary driving method and device based on multi-domain fusion control, vehicle and medium Technical Field The invention relates to the technical field of vehicle control, in particular to a garage auxiliary driving method and device based on multi-domain fusion control, a vehicle and a medium. Background The indoor garage generally adopts epoxy resin terrace paint, and when ground anti-tide or external rainwater is taken, the adhesion coefficient of the garage road surface is obviously reduced. The user runs on the wet and slippery ground of the garage, the adhesion coefficient is rapidly lower than that of a common scene, the speed of the vehicle is too high at the moment, or strong longitudinal force or transverse force is applied, so that the problems of vehicle slipping, tail flicking, head pushing and the like affecting safety are very easy to occur. Therefore, a method is needed to cope with such working conditions, and the safety of a driver in a wet slide garage is improved. Disclosure of Invention The invention provides a garage auxiliary driving method, device, vehicle and medium based on multi-domain fusion control, which are used for solving the problem of low driving safety of a driver in a wet slide garage. In a first aspect, the invention provides a garage auxiliary driving method based on multi-domain fusion control, which comprises the steps of judging whether a driving scene of a vehicle is located in a garage according to multi-mode information of the vehicle, and carrying out coordinated control on a braking system, a driving system and a suspension system of the vehicle when the driving scene is located in the garage so as to prevent slipping. According to the technical means, the garage scene is accurately identified through the multi-mode information, the linkage braking, driving and suspension systems are cooperatively controlled, and the problem of low adhesion coefficient of the wet road surface of the garage is solved in a targeted manner. Breaks through the control limitation of a single system, inhibits the risks of skidding, tail flicking and head pushing of the vehicle from multiple dimensions, does not need manual intervention of a driver, reduces the operation burden of the driver, and greatly improves the driving safety and stability of the wet slide garage. In some optional embodiments, judging whether the driving scene of the vehicle is located in the garage according to the multi-mode information of the vehicle comprises calculating sub-confidence scores of the driving scene belonging to the garage by utilizing each single-mode information of the vehicle, fusing each sub-confidence score to obtain a total confidence score, and determining that the driving scene of the vehicle is located in the garage when the total confidence score is greater than a preset confidence threshold. According to the technical means, the way of calculating the total confidence score by fusing the sub-confidence scores is adopted, so that the unilateralness and the misjudgment rate of single-mode information identification are avoided. Through multi-dimensional data complementation verification, accuracy and reliability of garage scene recognition are remarkably improved, accurate triggering and exiting of garage auxiliary driving functions are guaranteed, potential safety hazards caused by missed triggering are avoided, normal driving experience is prevented from being influenced by false triggering, and a solid scene judgment basis is provided for follow-up cooperative control. In some alternative embodiments, the sub-confidence scores of the driving scene belonging to the garage are calculated by utilizing all single-mode information of the vehicle respectively, wherein the first sub-confidence score is calculated by utilizing collected sign visual data, the sign visual data comprises a garage gate lifting rod, a height limiting mark and a speed limiting mark, the second sub-confidence score is calculated by utilizing light sensing signal intensity detected by the glass position of the vehicle, the third sub-confidence score is calculated on the basis of the intensity of a navigation positioning signal, the confidence score of the environment where the vehicle is located is calculated as a closed environment according to three-dimensional map data perceived by the surrounding environment of the vehicle body and is used as a fourth sub-confidence score, the number of perceived scribing parking spaces is counted, the fifth sub-confidence score is calculated according to the relation between the number of scribing parking spaces and a preset parking space threshold value, the road type where the current vehicle position is matched in an intelligent driving assistance system map is utilized, the sixth sub-confidence score is calculated according to the road type, the confidence score of the current vehicle position where the current vehicle is located in a tunnel is calculated in the i