IP Library › Granted Patent US 12,459,121
Granted Patent B2
US 12,459,121 · App. 17/553,859 · Granted Nov 4, 2025

Integrity and safety checking for robots

Inventors: Norbert Stoeffler (Graefelfing, DE); Kay-Ulrich Scholl (Malsch, DE); Fabian Oboril (Karlsruhe, DE); Yang Peng (Munich, DE)
Assignee: INTEL CORPORATION
B25J9/1674B25J9/1664B25J19/02G05B2219/39001G05B2219/50193
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,459,121
App. No.
17/553,859
Granted
Nov 4, 2025
Kind
B2
Abstract

Disclosed herein are systems, devices, and methods for efficiently checking the integrity of a robot system. The integrity-checking system may generate a predefined motion instruction for a robot, where the predefined motion instruction instructs the robot to perform a random movement at a test time. The random movement may be associated with an expected observation at the test time. The integrity-checking system may also determine a systematic failure based on a difference between the expected observation and a current observation of the robot at the test time. The current observation may be determined from received sensor data, and if the integrity-checking system detects a failure, it may stop the robot's motion or other mitigating instructions.

Claims (44)

1 . A device comprising:

a processor configured to:

generate a normal movement instruction for a robot, wherein the normal movement instruction instructs the robot to follow a trajectory at a movement speed, wherein the trajectory and the movement speed are associated with an expected optic flow along the trajectory;

generate a predefined motion instruction for the robot, wherein the predefined motion instruction instructs the robot to perform a random movement at a test time along the trajectory, wherein the random movement is associated with an expected observation at the test time;

determine the expected optic flow based on the normal movement instruction and a ground window defined by a width of the robot and a forward length along the trajectory of the robot;

determine a systematic failure based on:

a first difference between the expected observation and a current observation of the robot at the test time, wherein the current observation is determined from received sensor data; or

a second difference between the expected optic flow and a measured optic flow, wherein the measured optic flow is based on received optic sensor data; and

control movement of the robot based on the systematic failure.

2 . The device of claim 1 , further comprising a transmitter configured to transmit the predefined motion instruction to the robot and a receiver configured to receive the received sensor data indicative of the current observation of the robot at the test time.

3 . The device of claim 1 , wherein the systematic failure comprises at least one of a sensor failure of a sensor that provides the received sensor data, a manipulator failure in a manipulator of the robot that is configured to perform the random movement, a processing failure of the processor, a transmission failure in transmitting the predefined motion instruction to the robot.

4 . The device of claim 1 , wherein the random movement at the test time comprises a modification to a trajectory plan of a manipulator of the robot, wherein the trajectory plan comprises manipulator movement instructions for a work task of the robot.

5 . The device of claim 4 , wherein the modification to the trajectory plan comprises a modified path of the manipulator from a start position to a target position, wherein the modified path is different from a typical path of the manipulator defined by the trajectory plan.

6 . The device of claim 4 , wherein the modification to the trajectory plan comprises a modified joint configuration of the manipulator at a target position, wherein the modified joint configuration is different from a typical joint configuration of the manipulator at the target position defined by the trajectory plan.

7 . The device of claim 1 , further comprising a sensor, wherein the device is configured to receive the received sensor data from the sensor, wherein the sensor comprises at least one of a depth sensor, a camera, a radar, a light ranging and detection sensor, or an ultrasonic sensor.

8 . The device of claim 2 , wherein the receiver is further configured to receive second sensor data indicative of the current observation of the robot at the test time, wherein the second sensor data is associated with a second sensor and the received sensor data is associated with a first sensor, wherein the first sensor is different from the second sensor, wherein the processor is configured to determine a latency difference between the first sensor and the second sensor based on the received sensor data and the second sensor data.

9 . The device of claim 1 , wherein the random movement comprises a uniquely identifiable movement.

10 . The device of claim 1 , wherein the processor is configured to generate a second predefined motion instruction for a second robot, wherein the second predefined motion instruction instructs the second robot to perform the random movement at the test time, wherein the processor is configured to determine an execution delay based on a third difference between the current observation and a second current observation of a second robot, wherein the second current observation is based on received second sensor data indicative of the second current observation of the second robot at the test time.

11 . The device of claim 1 , wherein the random movement comprises a fault-specific motion pattern.

12 . The device of claim 2 , wherein the processor is configured to generate a second predefined motion instruction for a second robot, wherein the second predefined motion instruction comprises a second random movement for the second robot, is coordinated with the predefined motion instruction, and is associated with a second expected observation at a response time that is after the test time, wherein the receiver is further configured to receive second sensor data indicative of a second current observation of the second robot at the response time, wherein the processor is configured to determine the systematic failure further based on a second-fourth difference between the second expected observation and the second current observation.

13 . A device comprising:

a processor configured to:

generate a motion instruction for a robot, wherein the motion instruction instructs the robot to follow a trajectory at a movement speed, wherein the trajectory and movement speed are associated with an expected observation along the trajectory, wherein the expected observation comprises an expected optic flow of the trajectory;

determine the expected optic flow based on the motion instruction and a ground window defined by a width of the robot and a forward length along the trajectory of the robot;

determine a systematic failure based on a difference between the expected observation and a current observation of the robot along the trajectory, wherein the current observation comprises a measured optic flow for an image sequence of the trajectory determined from received optic data; and

control movement of the robot based on the systematic failure.

14 . The device of claim 13 , further comprising a transmitter configured to transmit the motion instruction to the robot and a receiver configured to receive the received optic data indicative of the current observation of the robot.

15 . The device of claim 13 , wherein the forward length comprises a safe braking distance needed for the robot to stop moving along the trajectory.

16 . A non-transitory computer readable medium, including instructions which, if executed, cause a processor to:

generate a normal movement instruction for a robot, wherein the normal movement instruction instructs the robot to follow a trajectory at a movement speed, wherein the trajectory and the movement speed are associated with an expected optic flow along the trajectory;

generate a predefined motion instruction for the robot, wherein the predefined motion instruction instructs the robot to perform a random movement at a test time along the trajectory, wherein the random movement is associated with an expected observation at the test time; and

determine the expected optic flow based on the normal movement instruction and a ground window defined by a width of the robot and a forward length along the trajectory of the robot;

determine a systematic failure based on:

a first difference between the expected observation and a current observation of the robot at the test time, wherein the current observation is determined from received sensor data; or

a second difference between the expected optic flow and a measured optic flow, wherein the measured optic flow is based on received optic sensor data; and

control movement of the robot based on the systematic failure.

17 . The non-transitory computer readable medium of claim 16 , wherein the instructions are further configured to:

cause a transmitter to transmit the predefined motion instruction to the robot; and

cause a receiver to receive the received sensor data indicative of the current observation of the robot at the test time.

18 . The non-transitory computer readable medium of claim 16 , wherein the systematic failure comprises at least one of a sensor failure of a sensor that provides the received sensor data, a manipulator failure in a manipulator of the robot that is configured to perform the random movement, a processing failure of the processor, a transmission failure in transmitting the predefined motion instruction to the robot.

19 . The non-transitory computer readable medium of claim 16 , wherein the random movement at the test time comprises a modification to a trajectory plan of a manipulator of the robot, wherein the trajectory plan comprises manipulator movement instructions for a work task of the robot.

20 . The non-transitory computer readable medium of claim 19 , wherein the modification to the trajectory plan comprises a modified path of the manipulator from a start position to a target position, wherein the modified path is different from a typical path of the manipulator defined by the trajectory plan.

21 . The non-transitory computer readable medium of claim 19 , wherein the modification to the trajectory plan comprises a modified joint configuration of the manipulator at a target position, wherein the modified joint configuration is different from a typical joint configuration of the manipulator at the target position defined by the trajectory plan.

22 . The non-transitory computer readable medium of claim 16 , wherein the instructions are further configured to cause the processor to receive the received sensor data from a sensor, wherein the sensor comprises at least one of a depth sensor, a camera, a radar, a light ranging and detection sensor, or an ultrasonic sensor.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Mar 10, 2022
From: STOEFFLER, NORBERT; SCHOLL, KAY-ULRICH; OBORIL, FABIAN; PENG, YANG
To: INTEL CORPORATION
Reel/Frame 059216/0464 →
Continuity (1)
Related Publication 20220105633A1 · Apr 7, 2022
References Cited (12)
US 10335962B1 · Rosenberg · 2019 [cited by examiner]
US 20200104611A1 · Ross · 2020 [cited by examiner]
US 20200331146A1 · Vu · 2020 [cited by examiner]
US 20240199333A1 · Zizka · 2024 [cited by examiner]
S. Thoduka, J. Gall and P. G. Plöger, “Using Visual Anomaly Detection for Task Execution Monitoring,” 2021 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Prague, Czech Republic, 2021, pp. 46… [cited by examiner]
X. Li, W. Chen, W. Zhang, X. Gao and L. Hao, “Acceleration of the Development for Motion Planning Algorithms Using V-REP,” 2019 WRC Symposium on Advanced Robotics and Automation (WRC SARA), Beijing, China, 2019, pp. 7-1… [cited by examiner]
T. T. Andersen, H. B. Amor, N. A. Andersen and O. Ravn, “Measuring and Modelling Delays in Robot Manipulators for Temporally Precise Control Using Machine Learning,” 2015 IEEE 14th International Conference on Machine Le… [cited by examiner]
Moveit, “Moving robots into the future”, https://moveit.ros.org/; 8 pages, retrieved on Nov. 1, 2021. [cited by applicant]
B. K. P. Horn et al., “Determining Optical Flow”, Massachusetts Institute of Technology, Artificial Intelligence Laboratory, A. I. Memo No. 572, Apr. 1980, 28 pages. [cited by applicant]
Dan et al., “Moving Object Tracking Method Based on Improved Lucas-Kanade Sparse Optical Flow Algorithm”, 2017, International Smart Cities Conference (ISC2), 2017, pp. 1-5. [cited by applicant]
L. Matthies et al., “Kalman Filter-based Algorithms for Estimating Depth from Image Sequences” International Journal of Computer, Vision, 3, 209-236, 30 pages, 1989. [cited by applicant]
Mobileye, “Responsibility-Sensitive Safety, A mathematical model for automated vehicle safety”, https://www.mobileye.com/responsibility-sensitive-safety/, retrieved on Jan. 6, 2022, 10 pages. [cited by applicant]