Detecting and responding to geometric changes to robots

1. A computer-implemented method, comprising:

obtaining, by one or more processors, a geometric model of an actual robot in a selected pose, wherein obtaining the geometric model comprises determining one or more bounded volumes containing one or more operational components of the robot, wherein each of the one or more bounded volumes encompasses multiple possible configurations of a given operational component of the one or more operational components;

obtaining, by one or more of the processors from an image sensor, image data of the actual robot in the selected pose;

comparing, by one or more of the processors, the image data to the geometric model to detect a geometric difference between the geometric model and the actual robot;

determining, by one or more of the processors, that the geometric difference causes one or more of the bounded volumes to be violated;

providing, by one or more of the processors, output indicative of the violation; and

based on the output:

altering a kinematic model of the actual robot and operating the actual robot based on the altered kinematic model; or

ceasing operation of the actual robot.

2. The computer-implemented method of claim 1 , further comprising updating, by the one or more processors, the geometric model in response to the detected geometric difference.

3. The computer-implemented method of claim 1 , wherein the output comprises a warning that the robot will potentially behave unpredictably due to the geometric difference.

4. The computer-implemented method of claim 1 , wherein the geometric model of the robot is based on a computer-aided design of the actual robot.

5. The computer-implemented method of claim 1 , wherein at least one of the one or more bounded volumes comprises a sphere.

6. The computer-implemented method of claim 1 , wherein at least one of the one or more bounded volumes comprises a circle.

7. The computer-implemented method of claim 1 , wherein the providing comprises providing the output in response to a determination, by one or more of the processors, that the geometric difference between the actual robot and the geometric model satisfies one or more thresholds.

8. The computer-implemented method of claim 1 , wherein the image data is obtained from a two-dimensional camera.

9. The computer-implemented method of claim 1 , wherein the image data is obtained from a three-dimensional vision sensor.

10. The computer-implemented method of claim 1 , wherein the image data is obtained from a stereographic camera.

11. The computer-implemented method of claim 1 , wherein the image data is obtained at least in part from a camera mounted on the robot.

12. The computer-implemented method of claim 1 , wherein obtaining the geometric model comprises setting the geometric model to conform to the selected pose.

13. A system comprising one or more processors, an image sensor, and memory, wherein the memory stores instructions that, in response to execution of the instructions by the one or more processors, cause the one or more processors to:

obtain a geometric model of a robot in a selected pose, wherein the geometric model includes one or more bounded volumes containing one or more operational components of the robot, wherein each of the one or more bounded volumes encompasses multiple possible configurations of a given operational component of the one or more operational components;

obtain, from the image sensor, image data of the actual robot in the selected pose;

compare the image data to the geometric model to detect a geometric difference between the geometric model and the actual robot;

determine that the geometric difference causes one or more of the bounded volumes to be violated;

provide output indicative of the violation; and

based on the output:

alter a kinematic model of the actual robot and operating the actual robot based on the altered kinematic model; or

cease operation of the actual robot.

14. The system of claim 13 , further comprising instructions to update the geometric model in response to the detected geometric difference.

15. At least one non-transitory computer-readable medium comprising instructions that, in response to execution of the instructions by one or more processors, cause the one or more processors to perform the following operations:

obtaining a geometric model of a robot in a selected pose, wherein obtaining the geometric model comprises determining one or more bounded volumes containing one or more operational components of the robot, wherein each of the one or more bounded volumes encompasses multiple possible configurations of a given operational component of the one or more operational components;

obtaining, from an image sensor, image data of the actual robot in the selected pose;

comparing the image data to the geometric model to detect a geometric difference between the geometric model and the actual robot;

determining, by one or more of the processors, that the geometric difference causes one or more of the bounded volumes to be violated;

providing output indicative of the geometric difference between the geometric model and the actual robot; and

based on the output:

altering a kinematic model of the actual robot and operating the actual robot based on the altered kinematic model; or

ceasing operation of the actual robot.

16. The at least one non-transitory computer-readable medium of claim 15 , wherein the geometric model of the robot is based on a computer-aided design of the actual robot.