Search

CN-116552540-B - Accelerator mistaken stepping prevention method, system and storage medium

CN116552540BCN 116552540 BCN116552540 BCN 116552540BCN-116552540-B

Abstract

The invention provides a method, a system and a storage medium for preventing mistaken stepping of an accelerator, and relates to the technical field of intelligent auxiliary driving, wherein the method comprises the steps of detecting data of a driver stepping on the accelerator pedal in real time, carrying out abnormal recognition on the data of the driver stepping on the accelerator pedal through a deep learning abnormal detection model, and monitoring whether an obstacle exists on a travelling route or not and whether collision risks exist between the vehicle and the obstacle or not; and judging whether the driver steps on the accelerator by mistake according to the abnormal data and the collision risk, if a confirmation signal of stepping on the accelerator by mistake is received, inhibiting the accelerator, not responding to acceleration, and finally, when the driver completely releases the accelerator or steps on the brake pedal, controlling the control function of the execution module to exit, thereby realizing more intelligent and more accurate identification of the action of stepping on the accelerator by mistake and avoiding or reducing the collision accident risk caused by stepping on the accelerator by mistake.

Inventors

  • YANG YAOYU

Assignees

  • 重庆长安汽车股份有限公司

Dates

Publication Date
20260505
Application Date
20230327

Claims (6)

  1. 1. The method for preventing the accelerator from being stepped on by mistake is characterized by comprising the following steps of: The accelerator pedal parameter monitoring module monitors the data of the driver for stepping on the accelerator pedal in real time, and transmits the monitored data of the driver for stepping on the accelerator pedal to the parameter abnormality identification module, wherein the data of the driver for stepping on the accelerator pedal comprises the stepping force, the stepping opening degree and the stepping opening degree change rate of the accelerator pedal, and the stepping opening degree change rate refers to the differentiation of the stepping opening degree to time; The parameter anomaly identification module carries out anomaly identification on the data of the driver stepping on the accelerator pedal by using a deep learning anomaly detection model, and comprises the following steps: Acquiring a data set of normal acceleration of a historical accelerator pedal; The method comprises the specific pretreatment process of preprocessing a data set normally accelerated by a historical accelerator pedal, namely, carrying out digital feature extraction on the data set normally accelerated by the historical accelerator pedal to obtain digital dimension data of the data set, and carrying out feature conversion on the digital dimension data; Constructing a deep learning anomaly detection model, training the constructed deep learning anomaly detection model by utilizing the preprocessed data set to obtain a trained deep learning anomaly detection model; The trained deep learning abnormality detection model is utilized to carry out deep learning abnormality monitoring on the data of the single depression of the accelerator pedal by the driver, and the recognition result of the data abnormality of the single depression of the accelerator pedal by the driver is obtained; Updating the recognition result of the data abnormality of the accelerator pedal to a data set; The collision risk monitoring module monitors whether an obstacle exists on a vehicle running route or not and whether collision risk exists between the vehicle and the obstacle or not in real time; the mistaken-stepping accelerator judging module judges whether the driver mistakenly steps on the accelerator according to the abnormality of the data of the driver stepping on the accelerator pedal and whether collision risks exist between the vehicle and the obstacle, and sends a signal for confirming mistaken stepping on the accelerator pedal to the control executing module; the control execution module receives a confirmation signal of mistakenly stepping on the accelerator, suppresses the accelerator and does not respond to acceleration; When the driver completely releases the accelerator or presses the brake pedal, the control function of the control execution module exits.
  2. 2. The accelerator pedal error prevention method according to claim 1, further comprising the steps of: the interactive early warning module receives a confirmation signal of mistakenly stepping on the accelerator, and warns to prompt a driver to mistakenly step on the accelerator.
  3. 3. The method of claim 1, wherein if the predicted time to collision between the vehicle and the obstacle is less than a threshold time to collision, the vehicle and the obstacle are at risk of collision, otherwise the vehicle and the obstacle are not at risk of collision.
  4. 4. The system is characterized by comprising an accelerator pedal parameter monitoring module, a parameter abnormality identifying module, a collision risk monitoring module, an accelerator pedal mistaken stepping judging module and a control executing module; The accelerator pedal parameter monitoring module is used for monitoring the data of the driver stepping on the accelerator pedal in real time and transmitting the monitored data of the driver stepping on the accelerator pedal to the parameter abnormality identification module; The parameter abnormality identification module is used for carrying out abnormality identification on the data of the driver stepping on the accelerator pedal by using the deep learning abnormality detection model; The collision risk monitoring module is used for monitoring whether an obstacle exists on a vehicle running route and whether collision risks exist between the vehicle and the obstacle in real time; The wrong-stepping accelerator judging module is used for judging whether the driver steps on the accelerator by mistake according to the abnormality of the data of the driver stepping on the accelerator pedal and whether the collision risk exists between the vehicle and the obstacle, and sending a signal for confirming the wrong stepping on the accelerator pedal to the control executing module; and the control execution module is used for receiving a confirmation signal of mistakenly stepping on the accelerator, inhibiting the accelerator and not responding to acceleration, and when the driver completely releases the accelerator or steps on the brake pedal, the control function of the control execution module exits.
  5. 5. The accelerator pedal error prevention system according to claim 4, wherein the parameter anomaly identification module comprises a data feature extraction unit, a data feature input unit, an anomaly detection model unit and a model parameter adjustment unit; the data feature extraction unit is used for extracting multidimensional features of the obtained data set of the normal acceleration of the historical accelerator pedal to obtain multidimensional feature data of the action of stepping on the accelerator pedal; the data characteristic input unit is used for inputting multidimensional characteristic data obtained by processing data into the constructed deep learning abnormal detection model; the abnormality detection model unit is used for carrying out abnormality detection on the multidimensional characteristic data to obtain a recognition result of data abnormality of the accelerator pedal; The model parameter adjusting unit is used for adjusting network parameters of the deep learning abnormal detection model in the training process.
  6. 6. A computer-readable storage medium, wherein a throttle false-stepping prevention program is stored on the computer-readable storage medium, and when the throttle false-stepping prevention program is executed by a processor, the operations of the throttle false-stepping prevention method according to any one of claims 1 to 4 are realized.

Description

Accelerator mistaken stepping prevention method, system and storage medium Technical Field The invention relates to the technical field of intelligent auxiliary driving, in particular to a method, a system and a storage medium for preventing mistaken stepping on an accelerator. Background Traffic accidents caused by the fact that a driver mistakenly uses an accelerator pedal as a brake pedal frequently occur, and huge damage is brought to life health and property of people, so that the problem of mistakenly stepping on the accelerator by the driver is an important safety problem which needs to be solved in intelligent auxiliary driving of an automobile. The prior patent literature discloses a device and a method for mistakenly stepping on and correcting an accelerator pedal, when the stepping opening of the accelerator pedal and the stepping speed of the accelerator pedal exceed a threshold value of the preset stepping speed, and meanwhile, the attention state of a driver is an attention distraction state, and a collision risk exists in a vehicle, the driver is judged to mistakenly step on the accelerator pedal, otherwise, the driver is judged not to mistakenly step on the accelerator pedal, on one hand, the device recognizes through the monitoring of the attention distraction state and the attention distraction state of the driver, but the attention state of the driver is uncertain, even if the driver is in the attention distraction state in a normal driving state, misjudgment or misjudgment is caused, the behavior of the driver to mistakenly step on the accelerator is judged, on the other hand, the judgment of the accelerator pedal parameter is carried out only through the comparison relation between the opening and the speed and the set threshold value, and the comprehensive judgment of the machine learning technology is not adopted from the historical data triggering of the normal stepping on the accelerator pedal of the driver, and the recognition of the mistaken stepping on the accelerator is serious accidental and the accelerator pedal is not determined. The prior patent literature also discloses a control method, a system and a storage medium for mistakenly stepping on an accelerator, when the opening increment of the accelerator pedal reaches a preset threshold value, and meanwhile, when an obstacle exists in the moving direction of an automobile and the speed meets the preset speed requirement, the power output of an engine is stopped to be regulated, but the patent only recognizes the obstacle exists in the moving direction of the automobile, does not distinguish whether collision risks exist between the automobile and the obstacle, and meanwhile, only depends on the opening increment of the accelerator pedal to judge whether mistakenly stepping on the accelerator, and does not fully refer to the accelerator stepping force, the opening of the accelerator pedal and the change rate of the opening of the accelerator pedal to comprehensively judge, and in addition, the control method does not relate to processing logic and man-machine interaction design during acceleration inhibition, and has limited influence on eliminating mistakenly stepping on the accelerator. Disclosure of Invention In order to solve the defects in the prior art, the invention aims to more intelligently and more accurately identify the abnormal behavior of mistakenly stepping on the accelerator caused by mistakenly taking the accelerator pedal as a brake pedal through the parameters of stepping on the accelerator pedal by a driver, so as to avoid or reduce the collision accident risk caused by mistakenly stepping on the accelerator. In order to achieve the technical effects, the technical scheme of the invention is as follows: an accelerator mistaken stepping prevention method comprises the following steps: The parameter monitoring module of the accelerator pedal monitors the data that the driver steps on the accelerator pedal in real time, and transmits the monitored data that the driver steps on the accelerator pedal to the parameter abnormality identification module, the parameter abnormality identification module carries out abnormality identification on the data that the driver steps on the accelerator pedal by using the deep learning abnormality detection model, and the collision risk monitoring module monitors whether an obstacle exists on a vehicle travelling path and whether collision risks exist between the vehicle and the obstacle in real time; the mistaken-stepping accelerator judging module judges whether the driver mistakenly steps on the accelerator according to the abnormality of the data of the driver stepping on the accelerator pedal and whether collision risks exist between the vehicle and the obstacle, and sends a signal for confirming mistaken stepping on the accelerator pedal to the control executing module; the control execution module receives a confirmation signal of mistakenly stepping on the accelerato