Method and System for Determining an Accurate Position of An Autonomous Vehicle

• US 20200133301A1
• Filed: 12/17/2018
• Published: 04/30/2020
• Est. Priority Date: 10/31/2018
• Status: Active Grant

• Current Assignee: Wipro Limited
• Original Assignee: Wipro Limited
• Inventors: Sarkar, Manas, Sunil Kumar, Balaji
• Granted Patent: US 10,983,530 B2
• Time in Patent Office: Days
• CPC Class Codes: G01S 17/06 Systems determining position data of a target
• CPC Class Codes: G01S 17/42 Simultaneous measurement of distance and other co-ordinates indirect measurement G01S17/46
• CPC Class Codes: G01S 17/86 Combinations of lidar systems with systems other than lidar, radar or sonar, e.g. with direction finders
• CPC Class Codes: G01S 17/89 for mapping or imaging
• CPC Class Codes: G05D 1/0088 characterized by the autonomous decision making process, e.g. artificial intelligence, predefined behaviours using knowledge based models G06N5/00
• CPC Class Codes: G05D 1/027 comprising intertial navigation means, e.g. azimuth detector inertial navigation G01C21/16; inertial navigation combined with non-inertial navigation instruments G01C21/165
• CPC Class Codes: G05D 1/0274 using mapping information stored in a memory device navigation using map-matching G01C21/30
• CPC Class Codes: G05D 1/0278 using satellite positioning signals, e.g. GPS

Abstract

The present disclosure discloses method and an Electronic Control Unit (ECU) (101) of autonomous vehicle for determining an accurate position. The ECU (101) determines centroid coordinate from Global Positioning System (GPS) points, relative to autonomous vehicle and identifies approximate location and orientation of vehicle on pre-generated map based on centroid coordinate and Inertial Measurement Unit (IMU) data. Distance and direction of surrounding static infrastructure is identified from location and orientation of autonomous vehicle based on road boundaries analysis and data associated with objects adjacent to autonomous vehicle. A plurality of lidar reflection reference points are identified within distance and direction of static infrastructure based on heading direction of autonomous vehicle. Position of lidar reflection reference points are detected from iteratively selected shift positions from centroid coordinate. Thereafter, ECU (101) corrects initial position of autonomous vehicle by adding centroid coordinate with selected shift position to determine accurate position of autonomous vehicle.

Claims

1. A method of determining an accurate position of an autonomous vehicle, the method comprising:

determining, by an Electronic Control Unit (ECU) (101) of the autonomous vehicle, a centroid coordinate from a plurality of Global Positioning System (GPS) points, relative to the autonomous vehicle, at stationary position for a predefined time;

identifying, by the ECU (101), an approximate location and orientation of the vehicle on a pre-generated map based on the centroid coordinate and Inertial Measurement Unit (IMU) data of the autonomous vehicle;

identifying, by the ECU (101), an approximate distance and direction of surrounding static infrastructure from the approximate location and orientation of the autonomous vehicle, based on an analysis of road boundaries in the pre-generated map and data associated with objects adjacent to the autonomous vehicle;

identifying, by the ECU (101), a plurality of lidar reflection reference points within the approximate distance and direction of the static infrastructure, based on heading direction of the autonomous vehicle, wherein the plurality of lidar reflection reference points are marked and joined a reference centre of the autonomous vehicle on an imaginary two-dimensional map;

detecting, by the ECU (101), position of the plurality of lidar reflection reference points from iteratively selected shift positions, at the approximate distance and direction from the centroid coordinate on the pre-generated map, within a predefined range and direction; and

correcting, by the ECU (101), initial position of the autonomous vehicle by adding the centroid coordinate with one of the selected shift positions, to determine the accurate position of the autonomous vehicle.

2. The method as claimed in claim 1, wherein determining the centroid coordinate from the plurality of Global Positioning System (GPS) points comprises:

performing, by the ECU (101), until a set of GPS points from the plurality of GPS points, collected for a period, are at a predefined threshold distance, the steps of

connecting, by the ECU (101), highest distant GPS points by each other by imaginary straight line and calculating one or more centre GPS points;

discarding, by the ECU (101), GPS points connecting with each other, and keeping the calculated centre points;

determining, by the ECU (101), the centroid coordinate by randomly selecting one of the sets of calculated centre points.

3. The method as claimed in claim 1 further comprising identifying vehicle heading direction with respect to lidar view based on elevation change of a first lidar ring from the autonomous vehicle.

4. The method as claimed in claim 1, wherein identifying the plurality of lidar reflection reference points comprises performing one or more searches by modifying range of predefined angles, in case no lidar reflection reference points are identified at the approximate distance and direction of the static infrastructure.

5. The method as claimed in claim 1, wherein the pre-generated map is a two-dimensional map.

6. The method as claimed in claim 1 further comprising joining the plurality of lidar reflection reference points with the reference centre of the autonomous vehicle to form a polygon reference shape.

7. The method as claimed in claim 1, wherein detecting the plurality of lidar reflection reference points comprises moving from the plurality of shift position with respect to the GPS centroid coordinate in the identified distance and direction on the imaginary two-dimensional map to trace a polygon reference shape.

8. An Electronic Control Unit (ECU) (101) of an autonomous vehicle for determining an accurate position of the autonomous vehicle, comprising:

a processor (109); and

a memory (107) communicatively coupled to the processor (109), wherein the memory (107) stores processor instructions, which, on execution, causes the processor (109) to;

determine of the autonomous vehicle, a centroid coordinate from a plurality of Global Positioning System (GPS) points, relative to the autonomous vehicle, at stationary position for a predefined time;

identify an approximate location and orientation of the vehicle on a pre-generated map based on the centroid coordinate and Inertial Measurement Unit (IMU) data of the autonomous vehicle;

identify an approximate distance and direction of surrounding static infrastructure from the approximate location and orientation of the autonomous vehicle, based on an analysis of road boundaries in the pre-generated map and data associated with objects adjacent to the autonomous vehicle;

identify a plurality of lidar reflection reference points within the approximate distance and direction of the static infrastructure, based on heading direction of the autonomous vehicle, wherein the plurality of lidar reflection reference points are marked and joined with a reference centre of the autonomous vehicle on an imaginary two-dimensional map;

detect position of the plurality of lidar reflection reference points from iteratively selected shift positions, at the approximate distance and direction from the centroid coordinate on the pre-generated map, within a predefined range and direction; and

correct initial position of the autonomous vehicle by adding the centroid coordinate with one of the selected shift positions, to determine the accurate position of the autonomous vehicle.

9. The ECU (101) as claimed in claim 8, wherein the processor (109) determines the centroid coordinate from the plurality of Global Positioning System (GPS) points by:

performing, until a set of GPS points from the plurality of GPS points, collected for a period, are at a predefined threshold distance, the steps of;

connecting highest distant GPS points by each other by imaginary straight line and calculating one or more centre GPS points;

discarding GPS points connecting with each other, and keeping the calculated centre points;

determining the centroid coordinate by randomly selecting one of the sets of calculated centre points.

10. The ECU (101) as claimed in claim 8, wherein the processor (109) identifies the vehicle heading direction with respect to lidar view based on elevation change of a first lidar ring from the autonomous vehicle.

11. The ECU (101) as claimed in claim 8, wherein the processor (109) identifies the plurality of lidar reflection reference points by performing one or more searches by modifying range of predefined angles, in case no lidar reflection reference points are identified at the approximate distance and direction of the static infrastructure.

12. The ECU (101) as claimed in claim 8, wherein the pre-generated map is a two-dimensional map.

13. The ECU (101) as claimed in claim 8 wherein the processor (109) joins the plurality of lidar reflection reference points with the reference centre of the autonomous vehicle to form a polygon reference shape.

14. The ECU (101) as claimed in claim 8, wherein the processor (109) detects the plurality of lidar reflection reference points by moving. from the plurality of shift position with respect to the GPS centroid coordinate in the identified distance and direction on the imaginary two-dimensional map to trace a polygon reference shape.