CN-122009197-A - In-vehicle passenger intelligent occupation detection method based on multi-sensor fusion

CN122009197ACN 122009197 ACN122009197 ACN 122009197ACN-122009197-A

Abstract

The invention discloses an intelligent occupancy detection method for passengers in a vehicle based on multi-sensor fusion, which comprises the steps of judging failure conditions of seat pressure sensors and visual detection sensor assemblies of a main driver and a non-main driver, obtaining an occupancy state and passenger attributes according to detection results of the seat pressure sensors and the visual detection sensor assemblies if the failure conditions are not generated, obtaining the occupancy state according to fastening states of a main driver safety belt or state signals of a brake pedal if the seat pressure sensors and the visual detection sensor assemblies of the main driver and the non-main driver are generated, obtaining the occupancy state and passenger attributes of the main driver and the non-main driver according to detection results of the sensors which are not generated if the failure conditions are generated, and obtaining no occupancy at corresponding positions of the non-main driver if the seat pressure sensors and the visual detection sensor assemblies at corresponding positions of the non-main driver are generated. The invention can improve the cabin occupation recognition precision and reliability by fusing the visual recognition scheme.

Inventors

CHEN GUOSHUAI
WU CHENGDONG
SHU ZHAOKUN
WEN MIN
ZHU ZHONGHUA

Assignees

安徽江淮汽车集团股份有限公司

Dates

Publication Date: 20260512
Application Date: 20260318

Claims (10)

1. An intelligent occupancy detection method for an occupant in a vehicle based on multi-sensor fusion is characterized by comprising the following steps: S1, judging fault conditions of a main driving seat pressure sensor, a seat pressure sensor at a non-main driving corresponding position and a visual detection sensor assembly; step S2, if neither the main driving seat pressure sensor nor the visual detection sensor assembly fails, step S3 is executed, if both the main driving seat pressure sensor and the visual detection sensor assembly fail, step S4 is executed, if neither the seat pressure sensor nor the visual detection sensor assembly fails, step S5 is executed, if both the main driving position and the seat pressure sensor at the non-main driving corresponding position fail, and the visual detection sensor assembly does not fail, step S6 is executed, if both the main driving position and the seat pressure sensor at the non-main driving corresponding position fail, and the visual detection sensor assembly fails, step S7 is executed, and if both the seat pressure sensor at the non-main driving corresponding position and the visual detection sensor assembly fail, step S8 is executed; s3, when a driver opens a main driving position of a vehicle door to sit on, a left front area controller collects signals of a main driving seat pressure sensor and simultaneously recognizes driver information through a visual detection sensor assembly; S4, the left front area controller obtains a main driving occupation state according to the fastening state of the main driving safety belt or a state signal of a brake pedal; S5, when a person in the non-main driving seat occupies the seat, the left front area controller obtains the occupied state and the passenger attribute of the non-main driving corresponding position according to the detection results of the visual detection sensor assembly and the pressure sensor of the corresponding position; S6, the left front area controller obtains the occupation state and the passenger attribute of the corresponding positions of the main driving and the non-main driving according to the detection result of the visual detection sensor assembly; s7, the front left area controller obtains the occupation states of the corresponding positions of the main driver and the non-main driver according to the detection result of the pressure sensor at the corresponding position; And S8, the front left area controller judges that the position corresponding to the non-main driving does not have a occupation.
2. The intelligent occupancy detection method for passengers in a vehicle based on multi-sensor fusion according to claim 1, wherein in the case that the main driving seat pressure sensor and the visual detection sensor assembly are not failed, the step S3 is that when the driver opens the door to sit on the main driving position, the front left area controller collects signals of the main driving seat pressure sensor and simultaneously recognizes driver information through the visual detection sensor assembly, and the front left area controller obtains the main driving occupancy state and the main driver attribute according to detection results of the main driving seat pressure sensor and the visual detection sensor assembly, and the method comprises the following steps: the main driving seat pressure sensor collects resistance change before and after the main driving seat position is pressed through an electric signal so as to detect whether a person or a heavy object occupies a seat; the main driving seat pressure sensor transmits the detection result to the left front area controller; The integrated brake system sends a state signal of the brake pedal stepping to the left front area controller through the CAN bus; The driver monitoring sensor and the in-cabin monitoring sensor collect person occupation images of the position of the main driving seat; The driver monitoring sensor and the in-cabin monitoring sensor transmit the acquired images to the cabin controller; The cabin area controller processes images acquired by the driver monitoring sensor and the cabin monitoring sensor to obtain an image recognition result, and transmits the image recognition result to the left front area controller through the Ethernet; The left front area controller judges the main driving occupation state according to the detection result of the main driving seat pressure sensor, the image identification result of the cabin area controller and the state signal of the stepping of the brake pedal; The front left area controller judges the attribute of the main driver according to the image recognition result of the cockpit area controller.
3. The intelligent occupancy detection method for an occupant in a vehicle based on multi-sensor fusion according to claim 2, wherein the determining, by the front left area controller, the main driving occupancy state according to a detection result of a main driving seat pressure sensor, an image recognition result of a cabin area controller, and a state signal of a brake pedal depression includes: if any one of the following conditions is met, judging that the main driving has occupied space, namely judging that the main driving has a person by the cabin domain controller and judging that the main driving has a person and a brake pedal is stepped down by the main driving seat pressure sensor; if any one of the following conditions is met, judging that the main driving has no occupation, namely, judging that the main driving is unmanned by the cabin domain controller, judging that the main driving is unmanned by the main driving pressure sensor, and not stepping on a brake pedal.
4. The intelligent occupancy detection method for an occupant in a vehicle based on multi-sensor fusion according to claim 1, wherein in the case that both the main driving seat pressure sensor and the visual detection sensor assembly fail, the step S4, the front left zone controller obtains the main driving occupancy state according to the fastening state of the main driving safety belt or the state signal of the brake pedal, and includes: The left front area controller collects the fastening state of the safety belt through an electric signal; The integrated brake system sends a state signal of the brake pedal stepping to the left front area controller through the CAN bus; and the left front area controller judges the main driving occupation state according to the main driving safety belt fastening state and the brake pedal state signal.
5. The intelligent occupancy detection method for an occupant in a vehicle based on multi-sensor fusion of claim 4, wherein the front left zone controller determines a primary occupancy state based on a primary seatbelt fastened state and a brake pedal state signal, comprising: If any one of the following conditions is met, judging that the main driver has occupied space, namely, the pressure sensor fails, the vision sensor fails, the main driver safety belt is buckled, and the brake pedal is stepped down; If the pressure sensor fails and the visual detection sensor fails, the main driving safety belt is not tied, and the brake pedal is not stepped down, judging that the main driving has no occupied space.
6. The intelligent occupancy detection method for an occupant in a vehicle based on multi-sensor fusion according to claim 1, wherein in the case that neither the seat pressure sensor nor the visual detection sensor assembly in the corresponding position of the non-primary driving fails, the step S5 is that when the non-primary driving seat occupies a seat, the front left area controller obtains the occupancy state and the occupant attribute in the corresponding position of the non-primary driving according to the detection results of the visual detection sensor assembly and the pressure sensor in the corresponding position, and the method comprises: the configuration words of the number of seats are stored in the left front area controller in advance so as to identify the number of seats of the corresponding vehicle type; If a seat at the corresponding position exists, the seat pressure sensor at the corresponding position acquires the resistance change before and after the corresponding seat position is pressed through an electric signal so as to detect whether a person or a heavy object occupies a seat; The non-main driving seat pressure sensor at the corresponding position sends the detection result to the area controller at the corresponding position; transmitting the detection result of the seat pressure sensor to a left front area controller through the Ethernet by the area controller at the corresponding position; The driver monitoring sensor and the in-cabin monitoring sensor collect person occupation images corresponding to the positions of the seats; The driver monitoring sensor and the in-cabin monitoring sensor transmit the acquired images to the cabin controller; The cabin area controller processes images acquired by the driver monitoring sensor and the cabin monitoring sensor to obtain an image recognition result, and transmits the image recognition result to the left front area controller through the Ethernet; The left front area controller judges the occupation state of the corresponding position according to the detection result of the seat pressure sensor at the corresponding position and the image identification result of the cabin area controller; and the left front area controller judges the passenger attribute of the corresponding position according to the image recognition result of the cockpit area controller.
7. The intelligent occupancy detection method for an occupant in a vehicle based on multi-sensor fusion according to claim 6, wherein the determining, by the front left area controller, the occupancy state of the corresponding position according to the detection result of the seat pressure sensor of the corresponding position and the image recognition result of the cabin controller, includes: If the cabin domain controller judges that a person exists or a seat pressure sensor at a corresponding position judges that a person exists, judging that the seat position occupies a place; if the cabin domain controller judges that no person exists and the pressure sensor judges that no person exists, judging that the seat position does not occupy a space.
8. The intelligent occupancy detection method for vehicle occupants based on multi-sensor fusion according to claim 1, wherein when the seat pressure sensors at the main driving position and the non-main driving corresponding position are failed and the visual detection sensor assembly is not failed, the step S6, the front left area controller obtains occupancy states and occupant attributes of the main driving and non-main driving corresponding positions according to the detection result of the visual detection sensor assembly, and the method comprises the following steps: The integrated brake system sends a state signal of the brake pedal stepping to the left front area controller through the CAN bus; at least one of a driver monitoring sensor and an in-cabin monitoring sensor collects person occupancy images of a primary driving seat position and a non-primary driving seat position; At least one of the driver monitoring sensor and the in-cabin monitoring sensor transmits the acquired image to the cabin controller; The cabin area controller processes images acquired by at least one of the driver monitoring sensor and the cabin monitoring sensor to obtain an image recognition result, and transmits the image recognition result to the left front area controller through the Ethernet; The left front area controller judges a main driving occupation state and a non-main driving occupation state according to an image recognition result of the cabin area controller and a state signal of stepping on a brake pedal; The front left area controller judges the attribute of the main driver and the attribute of the non-main driver according to the image recognition result of the cockpit area controller, The front left area controller judges the main driving occupation state and the non-main driving occupation state according to the image recognition result of the cabin area controller and the state signal of the stepping of the brake pedal, and comprises the following steps: if any one of the following conditions is met, judging that the main driver occupies a space, wherein the cabin domain controller judges that the main driver has a person and a brake pedal is stepped down; if the cabin domain controller judges that the main driver is unmanned and the brake pedal is not stepped on, judging that the main driver has no occupied space; If the cabin domain controller judges that the non-main driving corresponding position has a person, judging that the non-main driving corresponding position has a occupation; And if the cabin domain controller judges that no person exists at the position corresponding to the non-main driving, judging that the position corresponding to the non-main driving does not have a occupation.
9. The intelligent occupancy detection method for an occupant in a vehicle based on multi-sensor fusion according to claim 1, wherein in the case that the seat pressure sensors at the main driving position and the non-main driving corresponding position are not failed and the visual detection sensor assembly is failed, the step S7 of the front left area controller obtains occupancy states of the main driving and the non-main driving corresponding positions according to the detection result of the pressure sensors at the corresponding positions, including: The main driving seat pressure sensor collects resistance change before and after the main driving seat position is pressed through an electric signal so as to detect whether a person or a heavy object occupies the main driving position; the non-main driving seat pressure sensor at the non-main driving corresponding position acquires the resistance change before and after the seat at the non-main driving corresponding position is pressed through an electric signal so as to detect whether a person or a heavy object occupies the seat at the non-main driving corresponding position; the main driving seat pressure sensor transmits the detection result to the left front area controller; The non-main driving seat pressure sensor at the corresponding position sends the detection result to the area controller at the corresponding position; transmitting the detection result of the seat pressure sensor to a left front area controller through the Ethernet by the area controller at the corresponding position; The integrated brake system sends a state signal of the brake pedal stepping to the left front area controller through the CAN bus; and the left front area controller judges the main driving occupation state and the occupation state of the non-main driving corresponding position according to the detection results of the main driving seat pressure sensor and the non-main driving seat pressure sensor and the state signal of the stepping of the brake pedal.
10. The intelligent occupancy detection method for an occupant in a vehicle based on multi-sensor fusion according to claim 9, wherein the left front area controller judges the occupancy state of a primary driving and the occupancy state of a non-primary driving corresponding position according to detection results of a primary driving seat pressure sensor, a non-primary driving seat pressure sensor and a state signal of a brake pedal depression, and comprises: if any one of the following conditions is met, judging that the main driver occupies a space, wherein the main driver seat pressure sensor judges that a main driver has a person and a brake pedal is stepped down; if the main driving seat pressure sensor judges that the main driving is unmanned and the brake pedal is not stepped, judging that the main driving has no occupied space; If the non-main driving seat pressure sensor at the corresponding position judges that the non-main driving corresponding position has a person, judging that the non-main driving corresponding position has a occupation; And if the non-main driving seat pressure sensor at the corresponding position judges that the non-main driving corresponding position is unmanned, judging that the non-main driving corresponding position does not have a occupation.

Description

In-vehicle passenger intelligent occupation detection method based on multi-sensor fusion Technical Field The invention relates to the technical field of vehicle occupant occupation detection, in particular to an intelligent in-vehicle occupant occupation detection method based on multi-sensor fusion. Background Currently, the automobile industry is undergoing a drastic change of dynamism and intellectualization, and along with market evolution, the intelligent demands of users on new energy automobiles are increasing, especially in the aspect of intelligent scenes, such as convenient and comfortable functions of starting vehicles without sense, intelligent massage, intelligent partition air-conditioning, intelligent voice zones and the like. The realization of the functional scenes depends on the recognition result of the occupancy detection of the passengers in the vehicle. Currently, the mainstream occupation detection mode of industry adopts the mode of arranging pressure sensor below each seat cushion in the car, and when someone takes, pressure is exerted on the sensor generally, and the resistance value of sensor can change, and the controller is through detecting sensor resistance value change and is judged that the seat has someone to take. However, the detection mode has obvious defects that people can sit when heavy objects are placed on the seat, or people cannot be identified when partial passengers are lighter, so that functions such as intelligent air conditioning cannot be activated due to incapability of activating vehicle driving by a lighter female driver or incapability of identifying passengers by a rear row, and the like can not be detected completely when the pressure sensor fails, and the situation can cause complaints caused by poor user experience and further influence the image of a vehicle enterprise product. Therefore, there is a need for an intelligent occupant occupancy detection method based on multi-sensor fusion. Disclosure of Invention The invention aims to provide an intelligent occupancy detection method for passengers in a vehicle based on multi-sensor fusion, which is used for solving the problems in the prior art and can be used for solving the problems in the prior art. The invention provides an intelligent occupancy detection method for passengers in a vehicle based on multi-sensor fusion, which comprises the following steps: S1, judging fault conditions of a main driving seat pressure sensor, a seat pressure sensor at a non-main driving corresponding position and a visual detection sensor assembly; step S2, if neither the main driving seat pressure sensor nor the visual detection sensor assembly fails, step S3 is executed, if both the main driving seat pressure sensor and the visual detection sensor assembly fail, step S4 is executed, if neither the seat pressure sensor nor the visual detection sensor assembly fails, step S5 is executed, if both the main driving position and the seat pressure sensor at the non-main driving corresponding position fail, and the visual detection sensor assembly does not fail, step S6 is executed, if both the main driving position and the seat pressure sensor at the non-main driving corresponding position fail, and the visual detection sensor assembly fails, step S7 is executed, and if both the seat pressure sensor at the non-main driving corresponding position and the visual detection sensor assembly fail, step S8 is executed; s3, when a driver opens a main driving position of a vehicle door to sit on, a left front area controller collects signals of a main driving seat pressure sensor and simultaneously recognizes driver information through a visual detection sensor assembly; S4, the left front area controller obtains a main driving occupation state according to the fastening state of the main driving safety belt or a state signal of a brake pedal; S5, when a person in the non-main driving seat occupies the seat, the left front area controller obtains the occupied state and the passenger attribute of the non-main driving corresponding position according to the detection results of the visual detection sensor assembly and the pressure sensor of the corresponding position; S6, the left front area controller obtains the occupation state and the passenger attribute of the corresponding positions of the main driving and the non-main driving according to the detection result of the visual detection sensor assembly; s7, the front left area controller obtains the occupation states of the corresponding positions of the main driver and the non-main driver according to the detection result of the pressure sensor at the corresponding position; And S8, the front left area controller judges that the position corresponding to the non-main driving does not have a occupation. In the method for detecting the intelligent occupancy of the vehicle occupant based on the multi-sensor fusion, preferably, in the case that the main driving seat pressure senso