Control system and method for brake bleeding

• US 10,029,372 B2
• Filed: 12/11/2015
• Issued: 07/24/2018
• Est. Priority Date: 12/11/2015
• Status: Active Grant

• Current Assignee: GE Global Sourcing LLC (Westinghouse Air Brake Technologies Corporation)
• Original Assignee: General Electric Company
• Inventors: Tan, Huan, Lizzi, John Michael, Forman, Douglas, Theurer, Charles Burton, Al Assad, Omar, Castillo-Effen, Mauricio, Patrick, Romano, Kannan, Balajee, Gefen, Yonatan
• Primary Examiner(s): Oh, Harry Y
• Application Number:
  US14/966,717
• Publication Number:
  US 20170165839A1
• Time in Patent Office: 956 Days
• Field of Search: 700253
• CPC Class Codes: B25J 11/00 Manipulators not otherwise provided for
• CPC Class Codes: B25J 13/085 Force or torque sensors B25J13/082, B25J13/084 take precedence
• CPC Class Codes: B25J 15/00 Gripping heads and other end effectors grippers used in machine tools B23Q7/04; gripping members fitted on cranes B66C1/42, B66C1/44; gripping means used in the manufacture of semiconductors H01L21/68707; gripping means used for mounting electrical components H05K13/04
• CPC Class Codes: B25J 19/021 Optical sensing devices
• CPC Class Codes: B25J 9/1666 Avoiding collision or forbidden zones
• CPC Class Codes: B25J 9/1679 characterised by the tasks executed
• CPC Class Codes: B25J 9/1697 Vision controlled systems
• CPC Class Codes: B60T 15/048 Controlling pressure brakes of railway vehicles
• CPC Class Codes: B60T 15/54 for controlling exhaust from triple valve or from brake cylinder
• CPC Class Codes: B60T 7/16 operated by remote control, i.e. initiating means not mounted on vehicle
• CPC Class Codes: B61H 13/02 Hand or other personal actuation
• CPC Class Codes: G05B 2219/40039 Robot mounted or sliding inside vehicle, on assembly line or for test, service
• CPC Class Codes: G05B 2219/40586 6-DOF force sensor
• CPC Class Codes: G05B 2219/40613 Camera, laser scanner on end effector, hand eye manipulator, local
• CPC Class Codes: Y10S 901/01 Mobile robot

Abstract

A system includes a machine assembly, an imaging sensor, an encoder, and one or more processors. The machine assembly is movable to actuate a brake lever of a vehicle in order to open a valve of an air brake system of the vehicle. The imaging sensor acquires perception information of a working environment that includes the brake lever. The encoder detects a displaced position of the machine assembly relative to a reference position of the machine assembly. The one or more processors detect a position of the brake lever relative to the machine assembly based on the acquired perception information and the detected displacement of the arm. The one or more processors generate a motion trajectory for the machine assembly that provides a path to the brake lever. The one or more processors drive movement of the machine assembly along the motion trajectory towards the brake lever.

Claims

1. A system comprising:

a machine assembly that is movable to actuate a brake lever of a vehicle in order to open a valve of an air brake system of the vehicle;

an imaging sensor comprising an infrared emitter configured to generate and emit a pattern of infrared light and a depth camera configured to analyze the pattern of infrared light to interpret perceived distortions in the pattern to acquire three-dimensional perception information of a working environment that includes the brake lever;

an encoder detecting a displacement of the machine assembly relative to a reference position of a robotic arm of the machine assembly;

one or more processors configured to detect a position of the brake lever relative to the machine assembly based on the perception information and the displacement of the machine assembly, the one or more processors further configured to generate a motion trajectory for the machine assembly that provides a path from the machine assembly to the position of the brake lever, wherein the robotic arm moves along the motion trajectory towards the brake lever, the one or more processors further driving movement of the machine assembly along the motion trajectory towards the brake lever and revising the motion trajectory during movement of the robotic arm towards the brake lever with closed-loop feedback of the three-dimensional perception information and

analyze the three-dimensional perception information to construct a point cloud-based map of a working environment that includes the brake lever to identify a location of the brake lever, wherein the point cloud-based map comprises a first set of points associated with the brake lever, a second set of points associated with an obstacle, and a third set of points associated with free space.

2. The system of claim 1, further comprising a force sensor configured to operate with the machine assembly and detect one or more forces generated by the machine assembly as the machine assembly actuates the brake lever.

3. The system of claim 1, wherein the machine assembly includes a moveable arm that has an end-effector disposed at a distal end of the movable arm, the end-effector grasping the brake lever to actuate the brake lever, the imaging sensor being mounted on the movable arm at least one of on or proximate to the end-effector.

4. The system of claim 1, wherein the imaging sensor acquires the perception information at a frequency of at least 15 Hz.

5. The system of claim 1, wherein the one or more processors drive movement of the machine assembly at a control rate that is between 10 and 20 Hz.

6. The system of claim 1, wherein the one or more processors detect the position of the brake lever relative to the machine assembly at a detection rate, the one or more processors driving movement of the machine assembly at a control rate that is equal to the detection rate.

7. The system of claim 1, wherein the one or more processors are configured to detect the position of the brake lever relative to the machine assembly by performing two-dimensional image processing on the perception information acquired by the imaging sensor to identify at least one target area that potentially includes the brake lever, and then performing three-dimensional image segmentation on the at least one target area.

8. The system of claim 1, wherein the one or more processors are configured to analyze the perception information to detect the presence of one or more obstacles between the machine assembly and the brake lever, the one or more processors generating the motion trajectory such that the path for the machine assembly towards the brake lever avoids the one or more obstacles.

9. The system of claim 1, wherein the one or more processors generate the motion trajectory as a series of via-points between the machine assembly and the brake lever, the motion trajectory including dynamics values for position, velocity, and acceleration of the machine assembly associated with each via-point, the one or more processors configured to drive the machine assembly through the via-points with the dynamics values at each corresponding via-point.

10. The system of claim 1, wherein the machine assembly includes a movable arm having multiple joints and, subsequent to planning the motion trajectory, the one or more processors are configured to generate designated torques for corresponding joints of the arm, the one or more processors driving the arm of the machine assembly according to the designated torques.

11. The system of claim 10, further comprising a force sensor on the machine assembly that detects one or more forces generated by the machine assembly acting to actuate the brake lever, wherein the one or more processors are configured to revise the designated torques as the machine assembly is driven based on feedback received from the encoder and the force sensor.

12. The system of claim 1, wherein the imaging sensor acquires perception information corresponding to the position of the brake lever relative to the machine assembly as the machine assembly is driven towards the brake lever, the one or more processors configured to generate a revised motion trajectory for the machine assembly based on the perception information that is acquired as the machine assembly moves toward the brake lever.

13. A method comprising:

acquiring three-dimensional perception information to detect a position of a brake lever of a vehicle relative to an arm of a machine assembly, the brake lever being actuatable to open a valve of an air brake system of the vehicle, the perception information being acquired by an imaging sensor mounted on the arm by generating and emitting a pattern of infrared light using an infrared emitter and analyzing the pattern of infrared light to interpret perceived distortions in the pattern using a depth camera;

generating, using one or more processors, a motion trajectory for the arm from a current position of the arm to the position of the brake lever, the motion trajectory including a series of via-points that form a path for the arm to move along towards the brake lever while avoiding identified obstacles;

generating, using the one or more processors, designated torques for the arm at corresponding via-points based on the motion trajectory;

driving the arm towards the brake lever according to the designated torques to actuate the brake lever for opening the valve of the air brake system;

revising the motion trajectory during movement of the arm towards the brake lever with closed-loop feedback of the three-dimensional perception information; and

analyzing the three-dimensional perception information to construct a point cloud-based map of a working environment that includes the brake lever to identify a location of the brake lever, wherein the point cloud-based map comprises a first set of points associated with the brake lever, a second set of points associated with an obstacle, and a third set of points associated with free space.

14. The method of claim 13, further comprising receiving arm displacement information from an encoder as the arm moves according to the motion trajectory, and revising the designated torques based on the arm displacement information that is received, the arm displacement information including at least one of position information or motion information of the arm during movement of the arm towards the brake lever.

15. The method of claim 13, further comprising receiving force generation information from a force sensor as the arm actuates the brake lever, and revising the designated torques based on the force generation information that is received.

16. The method of claim 13, further comprising revising the motion trajectory based on updated perception information acquired by the imaging sensor during movement of the arm towards the brake lever.

17. The method of claim 13, wherein the position of the brake lever relative to the arm is detected by performing two-dimensional image processing on the perception information acquired by the imaging sensor to identify at least one target area that potentially includes the brake lever, and then performing three-dimensional image segmentation on the at least one target area without performing the three-dimensional image segmentation on one or more areas outside of the at least one target area.

18. A system comprising:

a machine assembly having an arm and a base, the arm being movable relative to the base to actuate a brake lever of a vehicle in order to open a valve of an air brake system of the vehicle;

an imaging sensor mounted on the arm, wherein the imaging sensor comprises an infrared emitter configured to generate and emit a pattern of infrared light and a depth camera configured to analyze the pattern of infrared light to interpret perceived distortions in the pattern to acquire three-dimensional perception information of a working environment that includes the brake lever;

an encoder detecting a displacement of the arm relative to a reference position of the arm;

one or more processors configured to detect a position of the brake lever relative to the arm based on the three-dimensional perception information and the displacement of the arm, the one or more processors further configured to generate a motion trajectory for the arm that provides a path from the arm to the position of the brake lever, the one or more processors driving movement of the arm along the motion trajectory towards the brake lever, and revising the motion trajectory during movement of the arm towards the brake lever with closed-loop feedback of the three-dimensional perception information and

analyze the three-dimensional perception information to construct a point cloud-based map of a working environment that includes the brake lever to identify a location of the brake lever, wherein the point cloud-based map comprises a first set of points associated with the brake lever, a second set of points associated with an obstacle, and a third set of points associated with free space.

19. The system of claim 18, wherein the arm includes an end-effector at a distal end of the arm, the end-effector configured to grasp the brake lever, the imaging sensor being mounted at least one of on or proximate to the end-effector.

20. The system of claim 18, further comprising a force sensor operably coupled to the machine assembly, the force sensor detecting one or more forces generated by the arm as the arm actuates the brake lever.