Autonomous Behaviors For A Remote Vehicle

• US 20130204465A1
• Filed: 02/05/2013
• Published: 08/08/2013
• Est. Priority Date: 07/14/2006
• Status: Active Grant

• Current Assignee: Flir Detection, Inc. (Teledyne Technologies Incorporated)
• Original Assignee: iRobot Corporation
• Inventors: Phillips, Emilie, Powers, Aaron, Shein, Andrew, Jamieson, Josef P., Sawyer, Tyson
• Granted Patent: US 8,577,517 B2
• Time in Patent Office: Days
• US Class Current: 701/2
• CPC Class Codes: G01C 21/20 Instruments for performing navigational calculations G01C21/24, G01C21/26 take precedence
• CPC Class Codes: G05D 1/0033 by having the operator tracking the vehicle either by direct line of sight or via one or more cameras located remotely from the vehicle
• CPC Class Codes: G05D 1/0274 using mapping information stored in a memory device navigation using map-matching G01C21/30
• CPC Class Codes: G05D 1/0278 using satellite positioning signals, e.g. GPS

Abstract

A method of operating a remote vehicle configured to communicate with an operator control unit (OCU) includes executing a click-to-drive behavior, a cruise control behavior, and a retro-traverse behavior on a computing processor. The click-to-drive behavior includes receiving a picture or a video feed and determining a drive destination in the received picture or video feed. The cruise control behavior includes receiving an absolute heading and velocity commands from the OCU and computing a drive heading and a drive velocity. The a retro-traverse behavior includes generating a return path interconnecting at least two previously-traversed waypoints of a list of time-stamped waypoints, and executing a retro-traverse of the return path by navigating the remote vehicle successively to previous time-stamped waypoints in the waypoints list until a control signal is received from the operator control unit.

Claims

1. A method of operating a remote vehicle configured to communicate with an operator control unit, the method comprising:

executing a click-to-drive behavior on a computing processor, the click-to-drive behavior comprising;

receiving a picture or a video feed from at least one imaging device of the remote vehicle; and

determining a drive destination in the received picture or video feed;

executing a cruise control behavior on the computing processor, the cruise control behavior comprising;

receiving an absolute heading command and a velocity command from the operator control unit, the absolute heading and velocity commands associated with a displayed picture or video feed on the operator control unit from the at least one imaging device of the remote vehicle; and

computing a drive heading and a drive velocity using the absolute heading and velocity commands, a vehicle position at a capture time of the displayed picture or video feed, and a vehicle velocity at the capture time of the displayed picture or video feed; and

executing a retro-traverse behavior on the computing processor when the remote vehicle ceases to receive a control signal from the operator control unit, the retro-traverse behavior comprising;

generating a return path interconnecting in reverse order of timestamps at least two previously-traversed waypoints of a list of time-stamped waypoints stored in non-transitory memory; and

executing a retro-traverse of the return path by navigating the remote vehicle successively to previous time-stamped waypoints in the waypoints list until a control signal is received from the operator control unit.

2. The method of claim 1, wherein the click-to-drive behavior further comprises:

receiving a selected destination pixel in the received picture or video feed from the operator control unit; and

computing coordinates of the destination point in a ground plan supporting the remote vehicle corresponding to the selected destination pixel.

3. The method of claim 1, wherein the click-to-drive behavior further comprises:

receiving a drive direction command from the operator control unit; and

computing coordinates of the destination point in a ground plan supporting the remote vehicle based on the received drive direction command.

4. The method of claim 1, wherein the cruise control behavior further comprises computing a travel vector using the drive heading and the drive velocity.

5. The method of claim 1, wherein the absolute heading and velocity commands are relative to the vehicle position and the vehicle velocity at the capture time of the displayed picture or video feed.

6. The method of claim 1, wherein the time-stamped waypoints are separated by at least a minimum difference in time and distance.

7. The method of claim 1, wherein the retro-traverse behavior further comprises excluding waypoints recorded while previously executing a retro-traverse.

8. The method of claim 7, wherein the retro-traverse behavior further comprises recording a list of start time and end time pairs for each retro-traverse.

9. The method of claim 8, further comprising excluding waypoints recorded between the start time and the end time pairs for each of the previously-executed retro-traverses.

10. The method of claim 1, wherein the retro-traverse behavior further comprises determining a look-ahead drive vector based on starting and ending waypoints of the return path to allow smooth navigation of the remote vehicle along the return path.

11. The method of claim 10, further comprising navigating the remote vehicle using the look-ahead vector, when the look-ahead vector lies within a predetermined distance of each of the waypoints along the return path.

12. The method of claim 11, further comprising avoiding taking shortcuts that bypass waypoints in the return path while navigating the remote vehicle using the look-ahead vector.

13. The method of claim 10, further comprising determining traversal of a waypoint while navigating the remote vehicle using the look-ahead vector by using a radius of interception of the waypoint.

14. The method of claim 10, further comprising determining whether the remote vehicle has passed one of the waypoints while navigating the remote vehicle using the look-ahead vector by using a perpendicular plane with respect to a drive direction.

15. The method of claim 1, wherein navigating the remote vehicle along the return path comprises selecting a forward or backward direction to navigate the remote vehicle along the return path based on a determination of which direction requires the least rotation.

16. The method of claim 1, wherein the retro-traverse behavior further comprises determining a position of the remote vehicle using odometry.

17. The method of claim 1, wherein the retro-traverse behavior further comprises determining a position of the remote vehicle comprises using a global positioning system, tying global positioning system data to odometry data, and using a Kalman filter to provide confidence in the position of the remote vehicle based on the odometry data and the global positioning system data.

18. The method of claim 1, wherein the retro-traverse behavior further comprises determining a position of the remote vehicle by maintaining a global map and placing coordinates of the remote vehicle within the global map.

19. The method of claim 18, wherein determining a position of the remote vehicle comprises maintaining a destination point within the global map and adjusting a heading of the remote vehicle to navigate the remote vehicle toward the destination point.

20. The method of claim 1, further comprising receiving a trigger activating an interactive behavior, the interactive behavior executing on a computing processor and allowing partial teleoperation of the mobile robot during execution of a non-interactive behavior, the interactive behavior being used to tune the non-interactive behavior.