CN-121989963-A - Vehicle control device

Abstract

The present invention relates to a vehicle control device that improves a technique of estimating a posture of an object. A vehicle control device (1) is provided with a control unit (10) and an imaging unit (12) for imaging the inside of a vehicle, wherein the control unit (10) acquires an image of the inside of the vehicle from the imaging unit (12), detects an object from the image using an object detection technique, calculates the activity level of the object from the image, estimates the depth difference between the depth of the hand from the imaging unit (12) to the object and the depth from the imaging unit (12) to the posture stabilization facility when at least a part of the hand of the object is located in a detection area including the posture stabilization facility and the activity level of the object is equal to or less than an activity level threshold, sets a flag indicating that the posture of the object is a safe posture to 'ON' when the depth difference is equal to or less than a depth difference threshold, and sets the flag to 'OFF' when the depth difference is greater than the depth difference threshold.

Inventors

• YAMAZAKI MASAYUKI

Assignees

• 丰田自动车株式会社

Dates

Publication Date

20260508

Application Date

20251029

Priority Date

20241105

Claims (5)

1. 1. A vehicle control device is provided with a control unit and an imaging unit for imaging the inside of a vehicle, The control unit performs the following processing: an image of the interior of the vehicle is acquired from the imaging unit, Using object detection techniques, objects are detected from the images, Calculating the activity level of the subject from the image, When at least a part of the hand of the subject is located in a detection area including a posture-stabilizing facility and the activity level of the subject is equal to or less than an activity level threshold, a depth difference between the depth from the imaging unit to the hand of the subject and the depth from the imaging unit to the posture-stabilizing facility is estimated, When the depth difference is equal to or less than a depth difference threshold, a flag indicating that the posture of the object is a safe posture is set to "ON", and when the depth difference is greater than the depth difference threshold, the flag is set to "OFF".
2. 2. The vehicle control apparatus according to claim 1, wherein, The control unit sets the flag to "OFF" when at least a part of the hand of the subject is not located in the detection area or when the activity amount of the subject is greater than the activity amount threshold.
3. 3. The vehicle control apparatus according to claim 1, wherein, The vehicle is an autonomous vehicle.
4. 4. The vehicle control apparatus according to claim 1, wherein, The gesture stabilization facility is a handrail or a sling.
5. 5. The vehicle control device according to any one of claims 1 to 4, wherein, The control unit allows the vehicle to be started when the flag is set to "ON" for all the objects in the vehicle, and does not allow the vehicle to be started when the flag is set to "OFF" for at least 1 object in the vehicle.

Description

Vehicle control device Technical Field The present disclosure relates to a vehicle control apparatus. Background Conventionally, a technique for estimating the posture of an object is known. For example, patent document 1 discloses a technique relating to a method and an apparatus capable of performing object recognition by machine learning and estimating a position and a posture at the same time. Prior art literature Patent literature Patent document 1 Japanese patent application laid-open No. 2019-029021 Disclosure of Invention Problems to be solved by the invention When the posture determination is utilized in the determination of allowing the vehicle to start, it is desirable to determine that the posture of the passenger in the vehicle is a safe posture, for example, a gripping posture such as a gripping armrest. The determination of the grip posture by image recognition is mainly performed using machine learning. However, in order to cope with changes in environmental characteristics such as patterns of gloves and armrests and interior trim of vehicles, a large amount of learning data and high calculation costs may be required for the determination by machine learning. The present disclosure, which has been completed in view of the above-described circumstances, aims to improve a technique of estimating the posture of an object. Means for solving the problems The vehicle control device according to one embodiment of the present disclosure includes a control unit and an imaging unit for imaging a vehicle, The control unit performs the following processing: an image of the interior of the vehicle is acquired from the imaging unit, Using object detection techniques, objects are detected from the images, Calculating the activity level of the subject from the image, When at least a part of the hand of the subject is located in a detection area including a posture-stabilizing facility and the activity level of the subject is equal to or less than an activity level threshold, a depth difference between the depth from the imaging unit to the hand of the subject and the depth from the imaging unit to the posture-stabilizing facility is estimated, When the depth difference is equal to or less than a depth difference threshold, a flag indicating that the posture of the object is a safe posture is set to "ON", and when the depth difference is greater than the depth difference threshold, the flag is set to "OFF". Effects of the invention According to one embodiment of the present disclosure, a technique of estimating a posture of an object can be improved. Drawings Fig. 1 is a block diagram showing a schematic structure of a system of one embodiment of the present disclosure. Fig. 2 is a flowchart showing the operation of the vehicle control device according to the present embodiment. (Symbol description) 1, A vehicle control device, 10, a control part, 12, a shooting part, 14, a communication part and 16, a storage part. Detailed Description Embodiments of the present disclosure are described below with reference to the accompanying drawings. (Outline of the present embodiment) With reference to fig. 1, an outline of a vehicle control apparatus 1 according to an embodiment of the present disclosure will be described. The vehicle control device 1 is an electronic device mounted on a vehicle, for example, a computer. The vehicle control device 1 uses an arbitrary object detection technique to detect an object from an image in the vehicle. The image may also be a still image or a moving image. The object is a passenger in the vehicle. A vehicle is any vehicle that can be ridden by more than 1 passenger, such as an automobile, bus, or section bus. The vehicle may be an autonomous vehicle capable of realizing autonomous driving with a class of 1 to 5 defined in SAE (Society of Automotive Engineers ). The vehicle may also be a manually driven vehicle with a level of 0. The vehicle may also be remotely monitored by a monitor outside the vehicle. The vehicle may also be MaaS (Mobility AS A SERVICE, travel as a service) special purpose vehicle. First, the outline of the present embodiment will be described, and the details will be described later. The vehicle control device 1 of the present embodiment includes a control unit 10 and an imaging unit 12 for imaging the inside of the vehicle. The control unit 10 acquires an image of the inside of the vehicle from the imaging unit 12. The control section 10 detects an object from an image using an object detection technique. The control unit 10 calculates the activity level of the subject from the image. The control unit 10 estimates a depth difference between the depth from the imaging unit 12 to the hand of the subject and the depth from the imaging unit 12 to the posture-stabilizing facility when at least a part of the hand of the subject is located in the detection area including the posture-stabilizing facility and the activity level of the subject is equal to or less tha