Method for detecting objects with a pivotable sensor device

• US 7,864,302 B2
• Filed: 02/01/2007
• Issued: 01/04/2011
• Est. Priority Date: 02/22/2006
• Status: Expired due to Fees

• Current Assignee: Siemens AG
• Original Assignee: Siemens AG
• Inventors: Soika, Martin, Pook, Sebastian
• Primary Examiner(s): Tarcza; Thomas H
• Assistant Examiner(s): Ratcliffe; Luke D
• Application Number:
  US12/224,312
• Publication Number:
  US 20090310117A1
• Time in Patent Office: 1,433 Days
• Field of Search: 356 301- 315, 356 401- 41, 356 501- 515, 356 6- 22, 356/28, 356/28.5
• US Class Current: 356/4.01
• CPC Class Codes: G01S 13/88 Radar or analogous systems specially adapted for specific applications electromagnetic prospecting or detecting of objects, e.g. near-field detection, G01V3/00
• CPC Class Codes: G01S 13/931 of land vehicles
• CPC Class Codes: G01S 17/93 for anti-collision purposes

Abstract

A scan sensor emits detection beams for detecting objects in a scanning area within a scanning plane. The position of the scanning plane is changed by pivoting the sensor device in a scanning area, to produce multiple detection planes. Detection points of objects in the surroundings of the sensor device are detected by the detection beams in the detection planes. Lines are extracted from the detection points of a respective detection plan. Measuring points are determined at the intersection points of the lines with one or more predetermined measuring planes. The measuring points in a respective measuring plane are classified into blocks and lines are extracted on a block basis from the measuring points of the blocks generated, as a result of which structures of objects in the measuring planes are determined.

Claims

1. A method for detecting objects with a pivotable sensor device having a scan sensor, comprising:

emitting detection beams from the scan sensor to detect objects in a scanning range within a scanning plane, while changing a position of the scanning plane by pivoting the sensor device in a swivel range, as a result of which a plurality of detection planes are produced;

detecting detection points of objects in surroundings of the sensor device using the detection beams in the detection planes;

extracting first lines from the detection points of a respective detection plane;

determining measuring points as intersection points of the lines with at least one predetermined measuring plane;

classifying the measuring points in the respective measuring plane into blocks by

for a measuring point to be classified, selecting another measuring point, from among those measuring points in already existing blocks that meet a predefined distance criterion, as having a shortest distance from the measuring point to be classified, where the predetermined distance criterion for a respective measuring point in a respective block is met if a first distance between the measuring point to be classified and the respective measuring point is shorter than a distance measure which depends on a second distance of the respective measuring point from an adjacent measuring point of the respective block, and

placing the measuring point to be classified, adjacent to the selected measuring point, in a block containing the selected measuring point; and

extracting second lines on a block basis from the measuring points in the blocks generated to determine structures of objects in the measuring planes.

2. The method as claimed in claim 1, wherein said placing places the measuring point to be classified behind the selected measuring point.

3. The method as claimed in claim 2, further comprising forming a new block for the measuring point which is to be classified, when no other measuring point can be selected.

4. The method as claimed in claim 2, further comprising forming a new block for the measuring point which is to be classified, when no blocks are yet present, by placing the measuring point to be classified and the other measuring point having the shortest distance to the measuring point to be classified.

5. The method as claimed in claim 2, wherein the predefined distance criterion for the respective measuring point in the respective block is met if the measuring point to be classified is at a third distance from a pair of adjacent measuring points in the respective block which is shorter than a fourth distance between the pair of adjacent measuring points multiplied by a factor that is at least equal to 1.

6. The method as claimed in claim 5, wherein the factor is less than 2 and substantially equal to 1.5.

7. The method as claimed in claim 6, wherein said placing of the measuring point to be classified into the block containing the selected measuring point is only if the shortest distance from the selected measuring point to the measuring point to be classified is less than a predetermined absolute measurement.

8. The method as claimed in claim 7, wherein said placing of the measuring point to be classified into the block containing the selected measuring point is only performed if a fifth distance between the measuring point to be classified and an approximated straight line of best fit through the measuring points of the block and the measuring point to be classified falls below a predetermined value.

9. The method as claimed in claim 8, wherein when the respective block has a single measuring point therein, the single measuring point is rated as meeting the predefined distance criterion.

10. The method as claimed in claim 9, wherein said classifying and extracting the second lines are performed concurrently with said emitting, detecting, extracting the first lines and determining, and when a new measuring point is determined, said classifying of the new point begins immediately.

11. The method as claimed in claim 10, further comprising, after the sensor device has panned from one end of the swivel range to an opposite end of the swivel range, checking a distance criterion to determine whether the blocks generated by said classifying can be combined to form bigger blocks.

12. The method as claimed in claim 11, wherein said extracting of the first and second lines is carried out using a divide-and-conquer method.

13. The method as claimed in claim 12, wherein the method is used in navigating a vehicle and the method further comprises:

comparing the second lines with third lines from previously determined objects stored in a memory; and

determining a position of the vehicle relative to at least one of the objects if the second and third lines match, thereby enabling navigation of the vehicle without a driver.

14. The method as claimed in claim 13, wherein said determining the position of the vehicle uses a Kalman filter.

15. The method as claimed in claim 14, wherein the detection planes extend in a substantially vertical direction and the sensor system is pivoted around a substantially vertical axis.

16. The method as claimed in claim 15, wherein the measuring planes extending in a substantially horizontal direction.

17. The method as claimed in claim 16, wherein the measuring planes are substantially above the sensor system.

18. The method as claimed in claim 17, wherein the swivel range covers substantially 90 degrees.

19. The method as claimed in claim 18, wherein the scanning range includes an angle range of at least 180 degrees and substantially 190 degrees.

20. The method as claimed in claim 19, wherein the scan sensor when pivoted is oriented in an upwards direction at an angle of inclination around a vertical swiveling axis.

21. The method as claimed in claim 20, wherein the angle of inclination is substantially 45 degrees.

22. The method as claimed in claim 21, wherein a laser scanner is used as the scan sensor.

23. A device for detecting objects, comprising:

a sensor device with a scan sensor that, when in operation, emits detection beams to detect objects in a scanning range within a scanning plane while changing a position of the scanning plane by pivoting the sensor device in a swivel range, as a result of which a plurality of detection planes are produced, and detection points of objects in surroundings of the sensor device are detected by the detection beams in the detection planes; and

a computer unit coupled to said sensor device and programmed to extract first lines from the detection points of a respective detection plane, determine measuring points as intersection points of the lines with at least one predetermined measuring plane and classify the measuring points in the respective measuring plane are classified into blocks by

for a measuring point to be classified, selecting another measuring point, from among those measuring points in already existing blocks that meet a predefined distance criterion, as having a shortest distance from the measuring point to be classified, where the predetermined distance criterion for a respective measuring point in a respective block is met if a first distance between the measuring point to be classified and the respective measuring point is shorter than a distance measure which depends on a second distance of the respective measuring point from an adjacent measuring point of the respective block, and

placing the measuring point to be classified, adjacent to the selected measuring point, in a block containing the selected measuring point, and

extracting second lines on a block basis from the measuring points in the blocks generated to determine structures of objects in the measuring planes.

24. A vehicle operating without a driver, comprising:

a chassis supported by at least one transport mechanism;

a memory, mounted on the chassis, storing representational lines of registered objects;

a sensor device, mounted on the chassis, with a scan sensor that, when in operation, emits detection beams to detect objects in a scanning range within a scanning plane while changing a position of the scanning plane by pivoting the sensor device in a swivel range, as a result of which a plurality of detection planes are produced, and detection points of objects in surroundings of the sensor device are detected by the detection beams in the detection planes;

at least one computer unit, mounted on the chassis and coupled to said sensor device, programmed to extract first lines from the detection points of a respective detection plane, determine measuring points as intersection points of the lines with at least one predetermined measuring plane and classify the measuring points in the respective measuring plane are classified into blocks by

for a measuring point to be classified, selecting another measuring point, from among those measuring points in already existing blocks that meet a predefined distance criterion, as having a shortest distance from the measuring point to be classified, where the predetermined distance criterion for a respective measuring point in a respective block is met if a first distance between the measuring point to be classified and the respective measuring point is shorter than a distance measure which depends on a second distance of the respective measuring point from an adjacent measuring point of the respective block, and

placing the measuring point to be classified, adjacent to the selected measuring point, in a block containing the selected measuring point,

extracting second lines on a block basis from the measuring points in the blocks generated to determine structures of objects in the measuring planes, and

navigating the vehicle without a driver by comparing the second lines extracted from the respective measuring plane with the representational lines of the registered objects in said memory.

25. The vehicle as claimed in claim 24, wherein the vehicle is one of an automatically operating fork lift truck and an automatically operating cleaning machine.