CN110015302B

CN110015302B - Automated vehicle system and method for changing from automatic mode to manual mode near a construction area

Info

Publication number: CN110015302B
Application number: CN201811554728.3A
Authority: CN
Inventors: D·阿加瓦尔; M·H·劳尔; B·R·希恩布兰德
Original assignee: Delphi Technologies LLC
Current assignee: Delphi Technologies LLC
Priority date: 2017-12-22
Filing date: 2018-12-19
Publication date: 2022-06-07
Anticipated expiration: 2038-12-19
Also published as: CN110015302A

Abstract

A system (10) and method (100) for operating an autonomous vehicle (12) includes providing a detector (22) and a controller circuit (28). The detector (22) is configured to detect a construction object (24). The controller circuit (28) is configured to determine that the host vehicle (12) is approaching the construction area (20) based on detection of the construction object (24) by the detector (22). The controller circuit (28) and method (100) are further configured to change control of the host vehicle (12) from an automatic mode (14) to a manual mode (16) in accordance with a determination that the host vehicle (12) is proximate to the construction area (20), the automatic mode (14) characterized by the controller circuit (28) operating the host vehicle (12), the manual mode (16) characterized by an occupant (18) of the host vehicle (12) operating the host vehicle (12).

Description

Automated vehicle system and method for changing from automatic mode to manual mode near a construction area

Technical Field

The present disclosure relates generally to systems and methods for operating an autonomous vehicle, and more particularly, to changing control of a host vehicle (host-vehicle) from an autonomous mode, characterized by a controller circuit operating the host vehicle, to a manual mode, characterized by an occupant of the host vehicle operating the host vehicle, when the host vehicle approaches a construction area.

Drawings

The invention will now be described, by way of example, with reference to the accompanying drawings, in which:

FIG. 1 is a schematic diagram of a system for operating an autonomous vehicle, according to one embodiment;

FIG. 2 is a scenario encountered by the system of FIG. 1 according to one embodiment; and

FIG. 3 is a flow diagram of a method of operating the system of FIG. 1, according to one embodiment.

Detailed Description

FIG. 1 shows a non-limiting example of a system 10 for operating an autonomous vehicle 12 (hereinafter referred to as a host vehicle 12). As used herein, the term "autonomous vehicle" may apply to situations when operating the host vehicle 12 in the autonomous mode 14 (i.e., fully autonomous mode), where what an occupant 18 (i.e., an operator) of the host vehicle 12 is to do to operate the host vehicle 12 may simply be to specify a destination. However, complete automation is not always necessary. It is contemplated that the teachings presented herein are useful when operating the host-vehicle 12 in a semi-automatic or manual mode 16, where the degree or level of automation may include maneuvering the host-vehicle 12, but may simply be providing audible or visual warnings to an operator who generally controls the steering system, accelerator, and brakes of the host-vehicle 12. For example, the system 10 may assist an operator (i.e., the occupant 18) in navigating through a construction area only as needed to avoid collisions and/or collisions with, for example, objects (e.g., construction obstacles or construction equipment).

System 10 includes a detector 22 configured or used to detect various objects, including but not limited to instances of construction object 24. As will be appreciated by those skilled in the art of perceptual sensors, the detector 22 may include or be formed by, but is not limited to, one or more instances of a camera, radar, lidar, or any combination thereof. The detector 22 may consist of or may include the following: an on-board detector 22A mounted on the host vehicle 12 and/or an off-board detector 22B located remotely from the host vehicle 12. That is, the information for determining the presence of an object in proximity to the host vehicle 12 may be provided exclusively by the on-board detector 22A, exclusively by the off-board detector 22B, or by a combination of the on-board detector 22A and the off-board detector 22B. The off-board detector 22B may be part of the infrastructure, such as a traffic camera.

FIG. 2 illustrates a non-limiting example of a scene 26 encountered by the host vehicle 12, which shows a non-limiting example of a construction object 24 that may be present at or near the construction area 20 and that may be detected by the detector 22. By way of example and not limitation, construction object 24 may include, but is not limited to: a reconfigurable information flag 24A that may be used to indicate, for example, an expected construction start date and end date; traffic cones 24B; a road block 24C; construction worker 24D (holding a shovel); a person 24E wearing a safety helmet and/or vest and/or holding both side stop/jog signs; and/or construction equipment 24F. Although not shown, it is contemplated that other objects disposed away from the construction area 20 may also indicate that the host-vehicle 12 is approaching the construction area; such as a forward construction sign or a detour sign.

Returning to fig. 1, the system 10 includes a controller circuit 28, which may be referred to elsewhere herein as the controller 28. The controller 28 may include a processor 30, such as a microprocessor or other control circuitry, such as analog and/or digital control circuitry, including an Application Specific Integrated Circuit (ASIC) for processing data, as will be apparent to those skilled in the art. Thereafter, unless otherwise noted, any reference to the controller 28 being configured for a capability will also be construed to imply that the processor 30 may also be configured for the same capability. Controller 28 or processor 30 may include an input 50 configured to receive a signal 52 from detector 22. The controller 28 may include a memory 32, i.e., a non-transitory computer-readable storage medium, including non-volatile memory, such as electrically erasable programmable read-only memory (EEPROM), for storing one or more routines, thresholds, and capture data. The memory 32 may be part of the processor 30, or part of the controller 28, or separate from the controller 28, such as a remote memory stored in the cloud.

One or more routines may be executed by controller 28 or processor 30 to perform steps to determine whether host vehicle 12 is at or near work area 20 based on signals received by controller 28, as described elsewhere herein. That is, controller 28 or processor 30 is configured to determine that host vehicle 12 is proximate to construction area 20 based on detection of one or more instances of construction object 24 by detector 22. As used herein, a host vehicle 12 is considered to be proximate a construction area 20 when it becomes preferable for an occupant 18 to operate (e.g., maneuver) the host vehicle 12 rather than the controller 28, i.e., it is preferable to operate the host vehicle 12 in the manual mode 16 rather than the automatic mode 14. For example, the host vehicle 12 may be determined to be approaching the construction area 20 based on or may be determined from a determination that the host vehicle 12 is closer than a threshold distance 34 (two hundred meters (200 meters)) from the construction area 20. That is, when the host vehicle 12 is less than two hundred meters (<200 meters) from the nearest edge of the construction area 20, the host vehicle 12 may be considered to be proximate to the construction area 20.

In accordance with a determination that the host-vehicle 12 is approaching the construction area 20, the controller 28 is configured to change the control of the host-vehicle 12 from the automatic mode 14, characterized by the controller circuit 28 manipulating and/or operating the accelerator/brake of the host-vehicle, to the manual mode 16, characterized by the occupant 18 of the host-vehicle 12 manipulating and/or operating the accelerator/brake of the host-vehicle. That is, if the controller 28 determines that the host-vehicle 12 is at or near the construction area 20, the controller 28 forces or attempts to force a change in the operating mode from the automatic mode 14 to the manual mode 16, whereby the occupant 18 (e.g., an operator) controls the host-vehicle 12 to navigate through or around the construction area 20. It is contemplated that, despite manual mode 16 being employed, safety features, such as automatic braking and/or other collision avoidance related features, may still be enabled on host vehicle 12. It is also contemplated that situations may arise where the occupant 18 is unable to control the host-vehicle 12, and some alternatives may be required to continue operating in a driver-less manner (i.e., the automatic mode 14). These alternatives for operating the host vehicle 12 when the occupant 18 is disabled have been proposed in other patent documents and will therefore not be discussed here.

In order to determine whether/when the host vehicle 12 is approaching or approaching a construction area, the system 10 needs to determine whether/when the object detected by the detector 22 is an instance of a construction object 24, i.e., the object is a construction object 24 and thus associated with a construction area. Accordingly, controller 28 or processor 30 may include an image processor 36, with image processor 36 configured to perform object association by comparing image 40 from detector 22 with one or more instances of stored image 38. As used herein, the image 40 may be compared to a photograph provided by a camera, a radar return map provided by a radar, a point cloud provided by a lidar, or a mixture/combination of any two or more of a photograph, a radar return map, and a point cloud. As will be appreciated by those skilled in the art of object classification, the stored images 38 may include thousands of images, each of which has been previously classified as being associated with an instance of an object that may be found at or near the construction area 20.

By way of example and not limitation, and as noted above, a construction object 24 may be one or more of a construction sign 24A, a traffic cone 24B, a barricade 24C, a construction worker 24D, a person holding a stop/jog sign, and/or construction equipment 24F (e.g., an excavator, crane, dump truck, grader, or vehicle deploying a traffic cone). For each of these examples of construction objects 24, there may be multiple instances in the stored image showing comparable instances of construction objects from multiple perspectives. It is also contemplated that image processor 36 may be configured to perform optical character recognition such that controller 28 may "read" construction marker 24A or a stop/jog marker held by person 24E.

System 10 may include a receiver 42 configured to receive construction broadcast 44 from a transmitter 46, where transmitter 46 may be operated by a government entity or a construction company operating at construction area 20. The receiver 42 and the transmitter 46 may each be part of a different transceiver, such as a Dedicated Short Range Communication (DSRC) transceiver or a cellular network transceiver. Construction broadcast 44 may include information regarding the date/day/time the construction was performed, information regarding detours so that construction area 20 may be avoided, and/or coordinates that define the boundaries of construction area 20.

The system 10 may include a digital map 48 that specifies the location and/or boundaries of the construction area. The digital map 48 may be located at the host vehicle 12, e.g., stored in the controller 28, and may be updated periodically (e.g., daily, hourly, etc.), or the digital map 48 may be stored in the cloud and may be accessed by the controller 28 in real-time. It is also contemplated that updates to the digital map 48 associated with the construction area 20 may be included in the construction broadcast 44. That is, controller circuit 28 may be configured to access, in real-time, a local update or local modification of digital map 48 that specifies the location of construction area 20.

The digital map 48 may include information about the position (i.e., coordinates) of the localization-object 54. As used herein, the located object 54 may be any object whose position is relatively permanent (e.g., not periodically changing and/or not easily moving) and preferably has characteristics that make detection and classification of the located object 54 reliable. Suitable examples of the positioning object 54 include, but are not limited to: buildings, bridges, lane markings, road/shoulder transitions, traffic signs, light poles, and reference objects (objects placed solely as positioning objects 54). The construction project may temporarily or permanently remove the positioning object 54 from the location indicated in the digital map 48. Accordingly, a determination may be made that the host vehicle 12 is proximate the construction area 20 based on or in accordance with a determination that the positioning object 54 indicated in the digital map 48 is not detected by the detector 22. Alternatively, the proximity of the host vehicle 12 to the construction area 20 may be determined based on or in accordance with a determination that the object position of the positioned object 54 detected by the detector 22 is not located at the map position of the positioned object 54 indicated in the digital map 48 (as the positioned object may have moved only a few meters due to construction).

FIG. 3 shows a non-limiting example of a method 100 for operating an autonomous vehicle (e.g., host vehicle 12). As will be described in greater detail below, the method 100 generally involves detecting the presence of a construction area 20 (fig. 1 and 2) near the host vehicle 12 (i.e., near the host vehicle 12), and transferring control of the host vehicle 12 from the controller 28 to the occupant 18 of the host vehicle 12. That is, rather than continuing to operate the host-vehicle 12 in the automatic mode 14 (i.e., the unmanned or fully autonomous operating mode) while traveling through the construction area 20, the control of operation of the host-vehicle 12 is transferred to the occupant 18, i.e., the operator, by changing the operation of the host-vehicle 12 to the manual mode 16.

Operating the host vehicle in the autonomous mode, step 105, may include operating the host vehicle 12 in the autonomous mode 14, which is generally characterized by the controller circuit 28 or processor 30 having control of the steering system and/or accelerator/brakes of the host vehicle 12.

At step 110, a construction broadcast is received, which may include receiving the construction broadcast 44 by the receiver 42. Construction broadcast 44 may include information about construction area 20, such as: the boundaries of the construction area 20, the proposed detours, possible delay times, and/or changes in the normal speed limit while traveling through the construction area 20.

At step 115, an object is detected, which may include detecting an instance of the construction object 24 with the detector 22. The detector 22 may detect objects that are not considered instances of the construction object 24, as well as objects that are instances of the construction object 24, i.e., objects associated with the construction area 20.

Is the object a location object, step 120? Which may include determining that the host vehicle 12 is proximate the construction area 20 based on a determination that the positioning object 54 indicated in the digital map 48 is not detected by the detector 22. The controller may access the digital map 48 to determine whether the digital map 48 indicates the presence of an instance of a located object 54, such as a bridge or building, in the vicinity. The controller 28 may then operate the detector to find the located object 54 indicated on the digital map 48. If the positioned object 54 indicated on the digital map 48 is not found (i.e., detected) by the detector because, for example, the construction project has removed the positioned object 54, then this may be an indication that the host vehicle is approaching the construction area 20.

At step 125, a digital map is accessed, which may include accessing a digital map 48, the digital map 48 specifying, for example and without limitation, boundaries or locations of the construction area 20. The digital map 48 may be stored at the host vehicle, such as in the memory 32, or remotely from the host vehicle 12 in the cloud, or a combination thereof to mix the digital map 48. The access may be, for example, to look for the presence of an object (e.g., locating object 54) detected by detector 22, or to search an area of a digital map around the current location of host vehicle 12 for instances of the object that may be used for location of host vehicle 12. The location at which the digital map is initially searched may be determined or indicated by a location detector 56 that receives location signals from satellites 58, as will be appreciated by those skilled in the art.

Step 130, is the object location? This may include determining that the host vehicle 12 is approaching the construction area 20 based on a determination that an object location (e.g., global positioning system or GPS coordinates) of the positioning object 54 detected by the detector 22 is not located at a map location of the positioning object 54 indicated in the digital map 48.

Step 135, are objects associated with the construction area? Which may include determining whether/when an object (i.e., construction object 24) is associated with construction area 20. The construction object 24 may be one or more of a construction sign, a traffic cone, a barricade, a construction worker, and construction equipment.

Determining an approaching construction area may include determining that the host vehicle 12 is approaching the construction area 20 based on the detection of the construction object 24 by the detector 22, step 140.

In step 145, is the host vehicle closer than a threshold distance? This may include determining that the host vehicle 12 is proximate to the construction area based on determining that the host vehicle is closer to the construction area 20 than the threshold distance 34 (e.g., the threshold distance 34 is 200 meters (200 meters)).

The change to manual mode may include changing control of the host vehicle 12 from the automatic mode 14 to the manual mode 16 based on a determination that the host vehicle 12 is proximate to the construction area 20, step 150.

Accordingly, a system 10 for operating an autonomous vehicle, a controller 28 for the system 10, and a method 100 of operating the system 10 are provided. The system 10 advantageously enables an occupant 18 of the host vehicle to maneuver or otherwise operate the host vehicle 12 in or through the work area 20, rather than enabling the host vehicle 12 to continue operating in the autonomous mode 14, as lane markings and/or positioned objects 54 on which the host vehicle 12 is typically controlled during autonomous mode operation may be unavailable or unreliable.

While the present invention has been described in accordance with its preferred embodiments, it is not so limited, but is only within the scope of the appended claims.

Claims

1. A system (10) for operating a host vehicle (12), the system (10) comprising:

a detector (22) configured to detect a construction object (24); and

a controller circuit (28) configured to determine that the host vehicle (12) is approaching a construction area (20) based on detection of a construction object (24) by the detector (22), and to change control of the host vehicle (12) from an automatic mode (14) to a manual mode (16) based on the determination that the host vehicle (12) is approaching the construction area (20), the automatic mode (14) characterized by the controller circuit (28) maneuvering the host vehicle (12) and the manual mode (16) characterized by an occupant (18) of the host vehicle (12) maneuvering the host vehicle (12),

Wherein the system (10) includes a digital map (48) that specifies the construction area (20), and

wherein the host vehicle (12) is determined to be proximate to the construction area (20) in accordance with a determination that the positioning object (54) indicated in the digital map (48) is not detected by the detector (22).

2. The system (10) in accordance with claim 1, wherein the host vehicle (12) is determined to be proximate to the construction area (20) in accordance with a determination that the host vehicle (12) is closer than a threshold distance (34) from the construction area (20).

3. The system (10) according to claim 2, wherein the threshold distance (34) is 200 meters (200 m).

4. The system (10) of claim 1, wherein the construction object (24) is associated with a construction area (20).

5. The system (10) according to any one of claims 1-4, wherein the construction object (24) is one or more of a construction sign (24A), a traffic cone (24B), a barricade (24C), a construction worker (24D), and construction equipment (24F).

6. The system (10) according to any one of claims 1-4, wherein the system (10) includes a receiver (42), the receiver (42) configured to receive a construction broadcast (44).

7. The system (10) in accordance with claim 1, wherein the host vehicle (12) is determined to be proximate to the construction area (20) in accordance with a determination that an object location of a positioning object (54) detected by the detector (22) is not located at a map location of the positioning object (54) indicated in the digital map (48).

8. The system (10) in accordance with claim 1, wherein the digital map (48) is stored remotely from the host vehicle (12), and the system (10) includes a transceiver configured to access the digital map (48).

9. A controller circuit (28) for operating a host vehicle (12), the controller circuit (28) comprising:

an input (50) configured to receive a signal (52) from a detector (22), the detector (22) configured to detect a construction object (24); and

a processor (30) configured to determine that the host vehicle (12) is approaching a construction area (20) based on detection of a construction object (24) by the detector (22), and to change control of the host vehicle (12) from an automatic mode (14) to a manual mode (16) based on determination that the host vehicle (12) is approaching the construction area (20), the automatic mode (14) characterized by the controller circuit (28) maneuvering the host vehicle (12) and the manual mode (16) characterized by an occupant (18) of the host vehicle (12) maneuvering the host vehicle (12),

wherein the controller circuit (28) is configured to access a digital map (48) that specifies the construction area (20), and

10. The controller circuit (28) of claim 9, wherein the host vehicle (12) is determined to be proximate to the construction area (20) in accordance with a determination that the host vehicle (12) is closer than a threshold distance (34) to the construction area (20).

11. The controller circuit (28) of claim 9 or 10, wherein the controller circuit (28) is configured to receive a construction broadcast (44).

12. A method (100) for operating a host vehicle (12), the method (100) comprising:

operating (105) the host vehicle (12) in an automatic mode (14), the automatic mode (14) characterized by a controller circuit (28) operating the host vehicle (12);

detecting (135) a construction object (24) with a detector (22);

determining (140) that the host vehicle (12) is approaching a construction area (20) in dependence of the detector (22) detecting the construction object (24); and

changing (150) control of the host vehicle (12) from the automatic mode (14) to a manual mode (16) in accordance with a determination that the host vehicle (12) is proximate to the construction area (20), the manual mode (16) characterized by an occupant (18) of the host vehicle (12) maneuvering the host vehicle (12),

wherein the method (100) comprises accessing (125) a digital map (48) specifying the construction area (20), and

13. The method (100) of claim 12, wherein the host vehicle (12) is determined to be proximate to the construction area (20) in accordance with a determination that the host vehicle (12) is closer than a threshold distance (34) from the construction area (20).

14. The method (100) according to claim 13, wherein the threshold distance (34) is 200 meters (200 m).

15. The method (100) according to any one of claims 12-14, wherein the construction object (24) is associated with the construction area (20).

16. The method (100) according to any one of claims 12-14, wherein the construction object (24) is one or more of a construction sign (24A), a traffic cone (24B), a barricade (24C), a construction worker (24D), and construction equipment (24F).

17. The method (100) according to any one of claims 12-14, wherein the method (100) includes receiving (110), by a receiver (42), a construction broadcast (44).

18. The method (100) of claim 12, wherein the host vehicle (12) is determined to be proximate to the construction area (20) in accordance with a determination that an object location of a positioned object (54) detected by the detector (22) is not located at a map location of the positioned object (54) indicated in the digital map (48).

19. The method (100) of claim 12, wherein the digital map (48) is stored remotely from the host vehicle (12), and the method (100) includes accessing the digital map (48).

20. A first device, comprising:

one or more processors (30);

a memory (32); and

one or more programs (105-150) stored in the memory (32), the one or more programs including instructions for performing the method (100) according to any of claims 12-19.

21. A non-transitory computer readable storage medium comprising one or more programs (105) for execution by one or more processors (30) of a first device, the one or more programs comprising instructions which, when executed by the one or more processors, cause the first device to perform the method (100) of any of claims 12-19.