Automated control of interactions between self-driving vehicles and animals

• US 9,481,367 B1
• Filed: 10/14/2015
• Issued: 11/01/2016
• Est. Priority Date: 10/14/2015
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Kundu, Ashish, Gordon, Michael S., Kozloski, James R., Malkin, Peter K., Pickover, Clifford A.
• Primary Examiner(s): Wong, Yuen
• Application Number:
  US14/883,058
• Time in Patent Office: 384 Days
• US Class Current: 1/1
• CPC Class Codes: B60W 10/04 including control of propulsion units
• CPC Class Codes: B60W 10/18 including control of braking systems
• CPC Class Codes: B60W 10/20 including control of steering systems
• CPC Class Codes: B60W 10/30 including control of auxiliary equipment, e.g. air-conditioning compressors or oil pumps
• CPC Class Codes: B60W 2420/403 Image sensing, e.g. optical camera
• CPC Class Codes: B60W 2554/402 Type
• CPC Class Codes: B60W 2554/4042 Longitudinal speed
• CPC Class Codes: B60W 2554/4044 Direction of movement, e.g. backwards
• CPC Class Codes: B60W 2710/081 Speed
• CPC Class Codes: B60W 2710/18 Braking system
• CPC Class Codes: B60W 2710/20 Steering systems
• CPC Class Codes: B60W 2710/30 Auxiliary equipments
• CPC Class Codes: B60W 2720/10 Longitudinal speed
• CPC Class Codes: B60W 2756/10 Involving external transmission of data to or from the vehicle
• CPC Class Codes: B60W 30/09 Taking automatic action to avoid collision, e.g. braking and steering
• CPC Class Codes: B60W 30/0956 the prediction being responsive to traffic or environmental parameters
• CPC Class Codes: B60W 40/02 related to ambient conditions
• CPC Class Codes: B60W 50/0097 Predicting future conditions
• CPC Class Codes: B60W 60/0015 specially adapted for safety
• CPC Class Codes: B60W 60/0027 using trajectory prediction for other traffic participants
• CPC Class Codes: G05D 1/0088 : characterized by the autonomous decision making process, e.g. artificial intelligence, predefined behaviours using knowledge based models G06N5/00

Abstract

A computer-implemented method causes a self-driving vehicle (SDV) to avoid a physical encounter with an animal. One or more sensors on a self-driving vehicle (SDV) capture a description of an environment around an animal that the SDV is approaching. A camera captures an image of the animal that the SDV is approaching. An SDV on-board computer on the SDV determines an animal type of the animal based on the captured image of the animal, and then predicts a direction and speed of movement of the animal based on the animal type of the animal and the captured description of the environment of the animal. The SDV on-board computer then determines a probability of a physical encounter between the SDV and the animal, and adjusts the speed and direction of the SDV accordingly in order to reduce a likelihood of the SDV striking the animal.

Claims

1. A computer-implemented method for causing a self-driving vehicle (SDV) to avoid a physical encounter with an animal, the computer-implemented method comprising:

capturing, by one or more sensors on the self-driving vehicle (SDV), a description of an environment around the animal that the SDV is approaching;

capturing, by a camera, an image of the animal that the SDV is approaching;

determining, by an SDV on-board computer, a species of the animal based on the captured image of the animal;

predicting, by the SDV on-board computer, a direction and speed of movement of the animal based on the determined species of the animal and the captured description of the environment around the animal;

determining, by the SDV on-board computer, a probability of a physical encounter (E) between the SDV and the animal exceeding a predefined confidence value (C) based on the predicted direction of movement of the animal and a current speed and direction of movement of the SDV;

in response to determining, by the SDV on-board computer and based on the predicted direction and speed of movement of the animal and the current speed and direction of the SDV, that E is greater than C, the SDV on-board computer instructing an SDV control processor on the SDV to direct SDV vehicular physical control mechanisms on the SDV to adjust the current speed and direction of movement of the SDV, thereby reducing a likelihood of the SDV striking the animal;

predicting, by the SDV on-board computer and based on the predicted direction and speed of movement of the animal, a length of time that the animal will block a pathway of the SDV; and

in response to the predicted length of time exceeding a predetermined value, stopping, by the SDV on-board computer, the SDV and turning off, by the SDV on-board computer, an engine in the SDV.

2. The computer-implemented method of claim 1, wherein the SDV is a first SDV, wherein the first SDV blocks a field of view of a second SDV such that a camera in the second SDV is unable to detect the animal, and wherein the computer-implemented method further comprises:

transmitting, from the SDV on-board computer in the first SDV to an SDV on-board computer in the second SDV, the image of the animal that the camera on the first SDV captured.

3. The computer-implemented method of claim 1, wherein the SDV is a first SDV, wherein the first SDV blocks a field of view of a second SDV such that a camera in the second SDV is unable to detect the animal, and wherein the computer-implemented method further comprises:

transmitting, from the SDV on-board computer in the first SDV to an SDV on-board computer in the second SDV, the predicted direction and speed of movement of the animal that was created by the SDV on-board computer in the first SDV.

4. The computer-implemented method of claim 1, wherein the SDV is a first SDV, wherein the first SDV leads a second SDV, and wherein the computer-implemented method further comprises:

generating, by the SDV on-board computer on the first SDV, a recommended navigation plan to avoid striking the animal; and

transmitting, from the first SDV to the second SDV, the recommended navigation plan.

5. The computer-implemented method of claim 1, wherein the SDV is a first SDV, wherein the first SDV leads a second SDV, and wherein the computer-implemented method further comprises:

generating, by the SDV on-board computer on the first SDV, a recommended navigation plan to avoid striking the animal; and

transmitting, from the SDV on-board computer on the first SDV to an SDV on-board computer on the second SDV, instructions to move the second SDV to a position that is within a predetermined distance from the first SDV and to follow the recommended navigation plan along with the first SDV.

6. The computer-implemented method of claim 1, further comprising:

retrieving, by one or more processors, animal profile information about the animal;

assigning, by the one or more processors, the animal to a cohort of animals that have crossed a roadway on which the SDV is traveling, wherein the animal shares more than a predetermined quantity of traits with members of the cohort of animals;

retrieving, by the one or more processors, historical data describing movement by members of the cohort of animals when crossing the roadway on which the SDV is traveling;

predicting, by the one or more processors, when the animal will reach a position on the roadway that the SDV is approaching based on the historical data describing the movement by the members of the cohort of animals while crossing the position on the roadway that the SDV is approaching; and

further adjusting, by the SDV control processor on the SDV, the current speed and direction of movement of the SDV according to the historical data describing the movement by the members of the cohort of animals when crossing the position on the roadway that the SDV is approaching.

7. The processor-implemented method of claim 1, further comprising:

receiving, by one or more processors, roadway sensor readings from multiple roadway sensors, wherein each of the multiple roadway sensors detects different current conditions of the roadway;

weighting, by the one or more processors, each of the roadway sensor readings for the different current conditions of the roadway;

summing, by the one or more processors, weighted roadway sensor readings for the different current conditions of the roadway;

determining, by the one or more processors, whether the summed weighted roadway sensor readings exceed a predefined level; and

in response to determining that the summed roadway weighted sensor readings exceed the predefined level, further adjusting, by the SDV control processor on the SDV, the current speed and direction of movement of the SDV in order to avoid the SDV striking the animal at the position on the roadway that the SDV is approaching.

8. The processor-implemented method of claim 1, further comprising:

receiving, by one or more processors, operational readings from one or more operational sensors on the SDV, wherein the one or more operational sensors detect a current state of equipment on the SDV;

detecting, by the one or more processors and based on received operational readings, a fault with the equipment on the SDV; and

in response to detecting the fault with the equipment on the SDV, further adjusting, by the SDV control processor on the SDV, the current speed and direction of movement of the SDV in order to avoid the SDV striking the animal at the position on the roadway that the SDV is approaching.

9. The computer-implemented method of claim 1, further comprising:

detecting, by the camera, a non-motorized vehicle that is affixed to the animal; and

further predicting, by the SDV on-board computer, the direction and speed of movement of the animal based on a presence of and characteristics of the affixed vehicle.

10. A computer program product for causing a self-driving vehicle (SDV) to avoid a physical encounter with an animal, the computer program product comprising a non-transitory computer readable storage medium having program code embodied therewith, the program code readable and executable by a processor to perform a method, the method comprising:

capturing, by one or more sensors on the self-driving vehicle (SDV), a description of an environment around the animal that the SDV is approaching;

capturing, by a camera on the SDV, an image of the animal that the SDV is approaching;

determining, by an SDV on-board computer, a species of the animal based on the captured image of the animal;

predicting, by the SDV on-board computer, a direction and speed of movement of the animal based on the determined species of the animal and the captured description of the environment around the animal;

determining, by the SDV on-board computer, a probability of a physical encounter (E) between the SDV and the animal exceeding a predefined confidence value (C) based on the predicted direction of movement of the animal and a current speed and direction of movement of the SDV;

in response to determining, by the SDV on-board computer and based on the predicted direction and speed of movement of the animal and the current speed and direction of the SDV, that E is greater than C, the SDV on-board computer instructing an SDV control processor on the SDV to direct SDV vehicular physical control mechanisms on the SDV to adjust the current speed and direction of movement of the SDV, thereby reducing a likelihood of the SDV striking the animal;

predicting, based on the predicted direction and speed of movement of the animal, a length of time that the animal will block a pathway of the SDV; and

in response to the predicted length of time exceeding a predetermined value, stopping, by the SDV on-board computer, the SDV and turning off, by the SDV on-board computer, an engine in the SDV.

11. The computer program product of claim 10, wherein the SDV is a first SDV, wherein the first SDV blocks a field of view of a second SDV such that a camera in the second SDV is unable to detect the animal, and wherein the method further comprises:

transmitting, from the SDV on-board computer in the first SDV to an SDV on-board computer in the second SDV, the image of the animal that the camera on the first SDV captured.

12. The computer program product of claim 10, wherein the SDV is a first SDV, wherein the first SDV blocks a field of view of a second SDV such that a camera in the second SDV is unable to detect the animal, and wherein the method further comprises:

transmitting, from the SDV on-board computer in the first SDV to an SDV on-board computer in the second SDV, the predicted direction and speed of movement of the animal that was created by the SDV on-board computer in the first SDV.

13. The computer program product of claim 10, wherein the SDV is a first SDV, wherein the first SDV leads a second SDV, and wherein the method further comprises:

generating, by the SDV on-board computer on the first SDV, a recommended navigation plan to avoid striking the animal; and

transmitting, from the SDV on-board computer on first SDV to an SDV on-board computer on the second SDV, instructions to move the second SDV to a position that is within a predetermined distance from the first SDV and to follow the recommended navigation plan along with the first SDV.

14. A system for causing a self-driving vehicle (SDV) to avoid a physical encounter with an animal, comprising:

one or more sensors on the self-driving vehicle (SDV) that capture a description of an environment around the animal that the SDV is approaching;

a camera on the SDV that captures an image of the animal that the SDV is approaching; and

a processor coupled to a computer readable memory and a non-transitory computer readable storage medium, wherein the processor retrieves program instructions from the non-transitory computer readable storage medium and stores the program instructions in the computer readable memory, wherein the program instructions, when executed by the processor via the computer readable memory, perform a method comprising;

determining, by an SDV on-board computer, a species of the animal based on the captured image of the animal;

predicting, by the SDV on-board computer, a direction and speed of movement of the animal based on the determined species of the animal and the captured description of the environment around the animal;

determining, by the SDV on-board computer, a probability of a physical encounter (E) between the SDV and the animal exceeding a predefined confidence value (C) based on the predicted direction of movement of the animal and a current speed and direction of movement of the SDV;

in response to determining, by the SDV on-board computer and based on the predicted direction and speed of movement of the animal and the current speed and direction of the SDV, that E is greater than C, the SDV on-board computer instructing an SDV control processor on the SDV to direct SDV vehicular physical control mechanisms on the SDV to adjust the current speed and direction of movement of the SDV, thereby reducing a likelihood of the SDV striking the animal;

predicting, based on the predicted direction and speed of movement of the animal, a length of time that the animal will block a pathway of the SDV; and

in response to the predicted length of time exceeding a predetermined value, stopping, by the SDV on-board computer, the SDV and turning off, by the SDV on-board computer, an engine in the SDV.

15. The system of claim 14, wherein the SDV is a first SDV, wherein the first SDV blocks a field of view of a second SDV such that a camera in the second SDV is unable to detect the animal, and wherein the method further comprises:

transmitting, from the SDV on-board computer in the first SDV to an SDV on-board computer in the second SDV, the image of the animal that the camera on the first SDV captured.

16. The system of claim 14, wherein the SDV is a first SDV, wherein the first SDV blocks a field of view of a second SDV such that a camera in the second SDV is unable to detect the animal, and wherein the method further comprises:

transmitting, from the SDV on-board computer in the first SDV to an SDV on-board computer in the second SDV, the predicted direction and speed of movement of the animal that was created by the SDV on-board computer in the first SDV.

17. The system of claim 14, wherein the SDV is a first SDV, wherein the first SDV leads a second SDV, and wherein the method further comprises:

generating, by the SDV on-board computer on the first SDV, a recommended navigation plan to avoid striking the animal; and

transmitting, from the SDV on-board computer on first SDV to an SDV on-board computer on the second SDV, instructions to move the second SDV to a position that is within a predetermined distance from the first SDV and to follow the recommended navigation plan along with the first SDV.