JP-7854974-B2 - Object detection system

JP7854974B2JP 7854974 B2JP7854974 B2JP 7854974B2JP-7854974-B2

Inventors

エムラエイキンシスボット
セイハンウチャル
樋口雄大
劉永康
尾口健太郎

Assignees

トヨタモーターエンジニアリングアンドマニュファクチャリングノースアメリカ，インコーポレイティド

Dates

Publication Date: 20260507
Application Date: 20231010
Priority Date: 20221011

Claims (20)

An object detection system, Processor and A memory that is communicably connected to the processor and stores instructions, wherein when an instruction is executed by the processor, the processor receives the instructions. The system receives priority assignments set by the vehicle occupants, the priority assignments correspond to the customization of at least one warning characteristic, and the at least one warning characteristic includes a distance threshold. To detect objects in the external environment of the vehicle, To classify the object for object classification and identification, Based on the object classification, the priority assignment is applied to the object. A memory that issues a warning according to at least one warning characteristic, and issues the warning when the distance from the vehicle to the object satisfies the distance threshold, An object detection system equipped with the following features.
The object detection system according to claim 1, wherein the priority assignment is set by the crew for the object classification, and the instruction further causes the processor to apply the priority assignment to the object when the object falls within the object classification.
The object detection system according to claim 2, wherein the object classification is at least one of trees, tree branches, speed bumps, debris, curbs, walls, barricades, vehicles, pedestrians, cyclists, and shopping carts.
The object detection system according to claim 1, wherein the priority assignment is set by the crew for a specific object, and the instruction further causes the processor to apply the priority assignment to the object if the object is the specific object.
The object detection system according to claim 1, wherein the priority assignment is set by the occupant for the object location, and the instruction further causes the processor to apply the priority assignment to the object when the object is at the object location.
The object detection system according to claim 1, wherein the priority assignment is a priority assignment within a range of priority assignments from low priority to high priority, and the distance threshold increases as the priority assignment increases from low priority to high priority.
The object detection system according to claim 1, wherein the instruction further causes the processor to create a database of historical data having inputs relating the object classification to priority assignments, and the instruction further causes the processor to predict the priority assignment of detected objects based on the historical data.
The object detection system according to claim 1, wherein the at least one warning characteristic includes at least one of a warning type, warning intensity, and warning location.
The object detection system according to claim 1, wherein the instruction further includes causing the processor to display the external environment of the vehicle in grayscale and issuing the warning, and displaying the object in color.
The above instruction further instructs the processor to: The object is detected using at least one monocular camera image acquired by a monocular camera mounted on the vehicle. The object detection system according to claim 1, wherein the distance from the vehicle to the object is determined using a depth map generated based on the at least one monocular camera image.
A method for operating an object detection system, wherein the method is Receiving a priority assignment set by the vehicle occupants, wherein the priority assignment corresponds to the customization of at least one warning characteristic, and the at least one warning characteristic includes a distance threshold, The detection of objects in the external environment of the vehicle, To classify and identify the object, the object is classified, Applying the priority assignment to the object based on the object classification , The method involves issuing a warning according to at least one warning characteristic, wherein the warning is issued when the distance from the vehicle to the object satisfies the distance threshold, Methods that include...
The method according to claim 11 , wherein receiving a priority assignment set by a vehicle occupant includes receiving a priority assignment set by the occupant for the object classification, and applying the priority assignment to the object includes applying the priority assignment to the object if the object falls within the object classification.
The method according to claim 12, wherein the object classification is at least one of trees, tree branches, speed bumps, debris, curbs, walls, barricades, vehicles, pedestrians, cyclists, and shopping carts.
The method according to claim 11, wherein receiving a priority assignment set by a vehicle occupant includes receiving a priority assignment set by the occupant for a specific object, and applying the priority assignment to the object includes applying the priority assignment to the object if the object is the specific object.
The method according to claim 11, wherein receiving a priority assignment set by a vehicle occupant includes receiving a priority assignment set by the occupant for an object location, and applying the priority assignment to the object includes applying the priority assignment to the object when the object is at the object location.
The method according to claim 11, wherein the priority assignment is a priority assignment within a range of priority assignments from low priority to high priority, and the distance threshold increases as the priority assignment increases from low priority to high priority.
To create a database of historical data having inputs that associate the aforementioned object classification with priority assignment, To predict the assignment of the priority of the detected object based on the historical data, The method according to claim 11, further comprising:
The method according to claim 11, wherein the at least one warning characteristic includes at least one of the following: warning type, warning intensity, and warning location.
The method according to claim 11, further comprising displaying the external environment of the vehicle in grayscale, and the issuing of the warning comprising displaying the object in color.
The object is detected using at least one monocular camera image acquired by a monocular camera mounted on the vehicle, The distance from the vehicle to the object is determined using a depth map generated based on the at least one monocular camera image, The method according to claim 11, further comprising:

Description

The embodiments disclosed herein relate to an object detection system for a vehicle, and more specifically, to an object detection system for a vehicle that can be customized by the vehicle's occupants. Some vehicles include an object detection system that detects one or more objects near the vehicle and provides a warning to the vehicle's occupants when the vehicle is close to an object. Such an object detection system may use various types of sensors to detect objects, such as cameras, sonar sensors, lidar sensors, and/or radar sensors, which can detect the presence of an object and determine its distance. In some mechanisms, the object detection system can display a live view of the vehicle's surroundings, including the object; for example, a live camera view may be displayed on the vehicle's user interface. Furthermore, the object detection system may provide the vehicle's occupants with information regarding the distance between the vehicle and the detected object. Embodiments of an object detection system for vehicles and a method for operating such a vehicle object detection system are disclosed herein. In one embodiment, an object detection system is disclosed. The object detection system includes a processor and a memory communicatively connected to the processor. The memory stores instructions, which, when executed by the processor, cause the processor to receive priority assignments set by the vehicle occupants. The priority assignments correspond to the customization of at least one warning characteristic, which includes a distance threshold. The instructions also cause the processor to detect objects in the vehicle's external environment and apply the priority assignments to the objects. The instructions further cause the processor to issue a warning according to at least one warning characteristic. According to the distance threshold, a warning is issued if the distance from the vehicle to the object meets the distance threshold. In another embodiment, a method for operating an object detection system is disclosed. This method includes receiving a priority assignment set by the vehicle occupants. The priority assignment corresponds to the customization of at least one warning characteristic, which includes a distance threshold. The method also includes detecting an object in the vehicle's external environment and applying the priority assignment to the object. The method further includes issuing a warning according to at least one warning characteristic. According to the distance threshold, a warning is issued if the distance from the vehicle to the object meets the distance threshold. These and other embodiments are described in further detail below. The various features, advantages, and other applications of this embodiment will become clearer by referring to the detailed description and drawings below. This figure shows an example of a vehicle equipped with an object detection system.This figure shows an example of a vehicle user interface that an occupant may use to customize one or more warning characteristics of an object detection system by assigning priorities to one or more classes of objects.This diagram illustrates an example where a vehicle backs up and exits a private road, and objects within the private road are detected by an object detection system.This figure shows an example of a warning issued by the object detection system regarding an object detected in Figure 3A.This figure illustrates an example where a vehicle is parked in a home garage, and objects within the garage are detected by an object detection system.This figure shows an example of assigning priority to objects detected in Figure 4A.Figure 4A shows an example of a warning issued by the object detection system regarding an object detected in the figure.This figure shows an example of assigning priority to object locations.This figure shows an example of a monocular camera image captured by a monocular camera mounted on a vehicle.This figure shows an example of a depth map generated based on the monocular camera image in Figure 5A.This figure shows an example of a monocular depth estimation system.This figure shows an example of how to operate an object detection system.This figure shows an example of how to predict object priority assignments based on historical data.Figure 10 shows an example of how to detect one or more objects using a depth map. This disclosure describes an object detection system for a vehicle. The object detection system is customizable by the vehicle occupant by assigning priorities to object classes, specific objects, and/or object locations. The occupant can set priority assignments using the vehicle's user interface, and these assignments correspond to customization of at least one warning characteristic, e.g., a distance threshold between the vehicle and the object. Upon detecting an object in the vehicle's external environment, the object detection system can classify the object and a