METHOD AND DEVICE FOR SAFEGUARDING A HAZARDOUS WORKING AREA OF AN AUTOMATED MACHINE

• US 20150049911A1
• Filed: 08/26/2014
• Published: 02/19/2015
• Est. Priority Date: 03/16/2012
• Status: Active Grant

• Current Assignee: PILZ GmbH & Company KG
• Original Assignee: PILZ GmbH & Company KG
• Inventors: DOETTLING, Dietmar, PILZ, Thomas, FRANK, Ruediger, HUFNAGEL, Mark
• Granted Patent: US 9,489,730 B2
• Time in Patent Office: Days
• US Class Current: 382/103
• CPC Class Codes: F16P 3/142 using image capturing devices
• CPC Class Codes: G06F 18/24 Classification techniques
• CPC Class Codes: G06T 2207/10012 Stereo images
• CPC Class Codes: G06T 2207/30164 Workpiece; Machine component
• CPC Class Codes: G06T 7/0004 Industrial image inspection
• CPC Class Codes: G06T 7/285 using a sequence of stereo image pairs
• CPC Class Codes: G06V 2201/06 Recognition of objects for industrial automation
• CPC Class Codes: G06V 40/103 Static body considered as a whole, e.g. static pedestrian or occupant recognition
• CPC Class Codes: G08B 21/02 Alarms for ensuring the safety of persons
• CPC Class Codes: G08B 21/22 responsive to presence or absence of persons

Abstract

A method and device for safeguarding a hazardous working area includes a camera system for generating successive 3-D images of the working area, and an evaluation unit comprising a fail-safe foreign object detector, a classifier, a person tracker and a comparator. The foreign object detector generates a first signal comprising a first item of position information, representative of the position of a foreign object in the protection zone. The classifier attempts to identify the foreign object as a person. The person tracker tracks an identified person using a series of current 3-D images and, after each new 3-D image, determines a second item of position information representing the current position of the identified person. If the position of the foreign object according to the first item of position information and the position of the identified person are different from one another, a control signal for stopping the machine is generated.

Claims

1. A method for safeguarding a hazardous working area of an automated machine, comprising the steps of:

providing at least one sensor unit which provides a respective current 3-D image of the working area at defined intervals of time,

defining at least one protection zone in the working area,

cyclically monitoring the protection zone using a fail-safe foreign object detector which generates a first signal on the basis of the respective current 3-D image and the defined protection zone if a foreign object is detected in the protection zone, the first signal comprising a first item of position information which is representative of a position of the foreign object in the protection zone,

cyclically classifying the foreign object in the protection zone using a classifier, the classifier generating a second signal if the foreign object is identified as a person, wherein the machine is operated on the basis of the first signal and the second signal,

tracking an identified person over a series of the respective current 3-D images using a person tracker, the person tracker determining, after each new 3-D image, a second item of position information which represents a current position of the identified person in the protection zone, and

comparing the first and second items of position information, wherein the machine is stopped if the position of the foreign object according to the first item of position information and the position of the identified person according to the second item of position information are different from one another.

2. The method of claim 1, wherein the machine is also stopped if the classifier does not identify the foreign object as a person, and wherein the machine is allowed to continue operation if the classifier identifies the foreign object as a person.

3. The method of claim 2, wherein the classifier has a personal database containing personal data which characterize a number of selected persons, and wherein the machine is only allowed to be operated if the classifier identifies the foreign object, using the personal data, as one of the selected persons.

4. The method of claim 2, wherein the classifier comprises an RFID reader which is used to read an RFID tag identifying a selected person.

5. The method of claim 1, wherein the classifier further comprises a parts database containing parts data which characterize a number of selected parts, and wherein the machine is continued to be operated if the classifier identifies the foreign object, using the parts data, as a part which is allowed to enter the protection zone.

6. The method of claim 1, wherein the classifier comprises a body part detector which is used to determine a position of body parts of the identified person.

7. The method of claim 1, wherein the classifier has a non-fail-safe design.

8. The method of claim 1, wherein the person tracker has a non-fail-safe design.

9. The method of claim 1, wherein the classifier is inactive until the foreign object detector detects a foreign object in the protection zone.

10. The method of claim 1, wherein the classifier is already active before the foreign object detector detects a foreign object in the protection zone.

11. The method of claim 1, wherein a minimum distance to the machine is defined, and wherein the machine is stopped if the identified person in the protection zone comes closer to the machine than the minimum distance.

12. The method of claim 11, wherein the minimum distance is varied depending on a current machine movement.

13. The method of claim 1, wherein the machine is operated with at least one of a reduced speed and a modified movement path if the second signal is present.

14. A device for safeguarding a hazardous working area of an automated machine, comprising:

a sensor unit configured to generate a respective current 3-D image of the working area at defined intervals of time,

a configuration unit for defining at least one protection zone in the working area, and

an evaluation unit which has a foreign object detector, a classifier, a person tracker, a comparator and an output stage,

wherein the foreign object detector has a fail-safe design and is configured to cyclically generate a first signal on the basis of a respective current 3-D image and on the basis of the defined protection zone if a foreign object is detected in the protection zone, the first signal comprising a first item of position information which is representative of a position of the foreign object in the protection zone, wherein the classifier is designed to cyclically generate a second signal if the foreign object is identified as a person,

wherein the person tracker is designed to track an identified person over a series of the respective current 3-D images and, after each new 3-D image, to determine a second item of position information which represents a current position of the identified person in the protection zone,

wherein the comparator is designed to compare the first and second items of position information and to generate a third signal on the basis thereof, and

wherein the output stage is designed to generate a control signal for stopping the machine on the basis of the second signal and on the basis of the third signal, the control signal being designed to stop the machine if the position of the foreign object according to the first item of position information and the position of the identified person according to the second item of position information are different from one another.

15. A computer program product comprising a non-transitory data storage medium storing program code which, when executed by a processor, causes a computer to carry out a method for safeguarding a hazardous working area of an automated machine, the method comprising the steps of:

providing at least one sensor unit which provides a respective current 3-D image of the working area at defined intervals of time,

defining at least one protection zone in the working area,

cyclically monitoring the protection zone using a fail-safe foreign object detector which generates a first signal on the basis of the respective current 3-D image and the defined protection zone if a foreign object is detected in the protection zone, the first signal comprising a first item of position information which is representative of a position of the foreign object in the protection zone,

cyclically classifying the foreign object in the protection zone using a classifier, the classifier generating a second signal if the foreign object is identified as a person, wherein the machine is operated on the basis of the first signal and the second signal, tracking an identified person over a series of the respective current 3-D images using a person tracker, the person tracker determining, after each new 3-D image, a second item of position information which represents a current position of the identified person in the protection zone, and

comparing the first and second items of position information, wherein the machine is stopped if the position of the foreign object according to the first item of position information and the position of the identified person according to the second item of position information are different from one another.