Method for determining the self-motion of a vehicle

• US 8,135,176 B2
• Filed: 12/23/2005
• Issued: 03/13/2012
• Est. Priority Date: 01/04/2005
• Status: Active Grant

• Current Assignee: Robert Bosch GmbH
• Original Assignee: Robert Bosch GmbH
• Inventors: Wuerz-Wessel, Alexander, Stein, Fridtjof
• Primary Examiner(s): AZARIAN, SEYED H
• Application Number:
  US11/792,208
• Publication Number:
  US 20090214077A1
• Time in Patent Office: 2,272 Days
• Field of Search: 382/100, 382103-107, 382/113, 382/120, 382/12, 382/123, 382/155, 382/162, 382/168, 382/170, 382/171, 382/172, 382/173, 382/181, 382/189, 382/193, 382/199, 382/209, 382/232, 382254-260, 382/274, 382/276, 382287-291, 382/305, 382/312, 701/46, 701/45, 701/32, 701/214, 701/1, 340/572.1, 340/539.11
• US Class Current: 382/107
• CPC Class Codes: B60W 2420/403 Image sensing, e.g. optical camera
• CPC Class Codes: B60W 40/10 related to vehicle motion
• CPC Class Codes: G01P 3/36 Devices characterised by the use of optical means, e.g. using infrared, visible, or ultraviolet light G01P3/68 takes precedence; gyrometers using the Sagnac effect, i.e. rotation-induced shifts between counter-rotating electromagnetic beams G01C19/64
• CPC Class Codes: G01S 11/12 using electromagnetic waves other than radio waves
• CPC Class Codes: G06T 7/246 using feature-based methods, e.g. the tracking of corners or segments
• CPC Class Codes: G06T 7/73 using feature-based methods

Abstract

A method and a device for determining the self-motion of a vehicle in an environment are provided, in which at least part of the environment is recorded via snapshots by an imaging device mounted on the vehicle. At least two snapshots are analyzed for determining the optical flows of image points, reference points that seem to be stationary from the point of view of the imaging device being ascertained from the optical flows. The reference points are collected in an observed set, new reference points being dynamically added to the observed set with the aid of a first algorithm, and existing reference points being dynamically removed from the observed set with the aid of a second algorithm.

Claims

1. A method for determining a self-motion of a vehicle in a surrounding environment, comprising:

recording a plurality of snapshots of at least part of the environment by an imaging device fixed in a selected position on the vehicle;

analyzing at least two snapshots for determining optical flows of image points, wherein reference points that appear to be stationary from the point of view of the imaging device are ascertained from the optical flows, and wherein the reference points are grouped in an observed set; and

dynamically changing the observed set, wherein new reference points are dynamically added to the observed set with the aid of a first predetermined algorithm, and existing reference points are dynamically removed from the observed set with the aid of a second predetermined algorithm, and wherein the second predetermined algorithm includes at least one step for ascertaining permanently present reference points having unchanged positions.

2. The method as recited in claim 1, wherein the first predetermined algorithm includes at least one step for ascertaining image points whose optical flow is substantially zero.

3. The method as recited in claim 1, wherein the first predetermined algorithm includes at least one step for ascertaining image points whose optical flow corresponds to optical flow ascertained for a majority of reference points.

4. The method as recited in claim 1, wherein the first predetermined algorithm includes at least one step for filtering image points with the aid of a Kalman filter.

5. The method as recited in claim 1, wherein the first predetermined algorithm includes at least one step for ascertaining at least one object group from locally contiguously positioned reference points.

6. The method as recited in claim 1, wherein the second predetermined algorithm includes steps for ascertaining a first optical flow of a majority of reference points with respect to at least one second optical flow of a minority of reference points.

7. The method as recited in claim 1, wherein the second predetermined algorithm includes at least one step for filtering the reference points with the aid of a Kalman filter.

8. The method as recited in claim 1, wherein at least some of the reference points of the observed set are stored at different points in time, and wherein a curve of the reference points is determined from stored data.

9. The method as recited in claim 1, wherein an angular self-motion of one of the imaging device or the vehicle is calculated from an identical optical flow of a majority of reference points.

10. The method as recited in claim 9, wherein the ascertained angular self-motion is relayed to at least one control unit of the vehicle, and wherein the ascertained angular self-motion is taken into account by the at least one control unit in at least one of determining and executing a control intervention.

11. The method as recited in claim 9, wherein the ascertained angular self-motion is taken into account in the analysis of the optical flows such that the optical flows are filtered of a component that depends on the angular self-motion, whereby optical flows caused by substantially purely translational motion are obtained.

12. The method as recited in claim 1, wherein an angular self-motion of one of the imaging device or the vehicle is described using a mathematical system of equations, and wherein coefficients for calculating the angular self-motion are ascertained by filtering the ascertained observed set in the time range.

13. A device for determining a self-motion of a vehicle in a surrounding environment, comprising:

an imaging device for recording a plurality of snapshots of at least part of the environment, wherein the imaging device is fixed in a selected position on the vehicle;

an ascertaining device for determining optical flows of image points by analyzing at least two snapshots, wherein reference points that appear to be stationary from the point of view of the imaging device are ascertained from the optical flows, and wherein the reference points are grouped in an observed set;

a memory device for storing the observed set of the reference points; and

dynamically changing the observed set, wherein new reference points are dynamically added to the observed set with the aid of a first predetermined algorithm, and existing reference points are dynamically removed from the observed set with the aid of a second predetermined algorithm;

a first logic circuit and a second logic circuit for dynamically changing the observed set, wherein new reference points to be dynamically added to the observed set are determined with the aid of a first predetermined algorithm implemented by the first logic circuit, and wherein existing reference points to be dynamically removed from the observed set are determined with the aid of a second predetermined algorithm implemented by the second logic circuit.