APPARATUS AND METHOD FOR PROVIDING LOCATION AND HEADING INFORMATION OF AUTONOMOUS DRIVING VEHICLE ON ROAD WITHIN HOUSING COMPLEX

• US 20150142248A1
• Filed: 10/21/2014
• Published: 05/21/2015
• Est. Priority Date: 11/20/2013
• Status: Abandoned Application

• Current Assignee: Electronics and Telecommunications Research Institute
• Original Assignee: Electronics and Telecommunications Research Institute
• Inventors: HAN, Seung-Jun, SUNG, Kyung-Bok, CHOI, Jeong-Dan, MIN, Kyoung-Wook
• Application Number:
  US14/519,288
• Publication Number:
  US 20150142248A1
• Time in Patent Office: Days
• US Class Current: 701/23
• CPC Class Codes: G01C 21/3602 Input other than that of destination using image analysis, e.g. detection of road signs, lanes, buildings, real preceding vehicles using a camera

Abstract

Disclosed herein is an apparatus and method for providing location and heading information of an autonomous driving vehicle on a road within a housing complex. The apparatus includes an image sensor installed on an autonomous driving vehicle and configured to detect images of surroundings depending on motion of the autonomous driving vehicle. A wireless communication unit is installed on the autonomous driving vehicle and is configured to receive a Geographic Information System (GIS) map of inside of a housing complex transmitted from an in-housing complex management device in a wireless manner. A location/heading recognition unit is installed on the autonomous driving vehicle, and is configured to recognize location and heading of the autonomous driving vehicle based on the image information received from the image sensor and the GIS map of the inside of the housing complex received via the wireless communication unit.

Claims

1. An apparatus for providing location and heading information of an autonomous driving vehicle on a road within a housing complex, comprising:

an image sensor installed on an autonomous driving vehicle and configured to detect images of surroundings depending on motion of the autonomous driving vehicle;

a wireless communication unit installed on the autonomous driving vehicle and configured to receive a Geographic Information System (GIS) map of inside of a housing complex transmitted from an in-housing complex management device in a wireless manner; and

a location/heading recognition unit installed on the autonomous driving vehicle, and configured to recognize location and heading of the autonomous driving vehicle based on the image information received from the image sensor and the GIS map of the inside of the housing complex received via the wireless communication unit.

2. The apparatus of claim 1, wherein the location/heading recognition unit estimates ego-motion from stereoscopic images, acquired by the image sensor, using a multi-view image processing technique, converts the images collected by the image sensor into images projected onto a ground surface of the road to extract markers on the road, obtains a probability of matching the GIS map to estimate a current location and attitude, and estimates a dynamic behavioral state of the vehicle based on the estimated ego-motion and the estimated current location and attitude to output predicted results of a state of the vehicle.

3. The apparatus of claim 2, wherein the location/heading recognition unit is configured to:

define three-dimensional (3D) coordinates of feature points detected from an image initially acquired by the image sensor as initial values of a point cloud, and

track feature points acquired from a previous key frame on an image plane whenever each image is input from the image sensor, obtain relative coordinates at a current time based on a key frame using a Perspective-Three-Point (P3P) technique, and calculate the ego-motion from relative coordinates of the previous key frame and relative coordinates at the current time.

4. The apparatus of claim 3, wherein the location/heading recognition unit is configured such that, when movement of the autonomous driving vehicle above a predetermined amount occurs, and generation of a new key frame is required, a separate thread performs bundle adjustment and then updates the point cloud.

5. The apparatus of claim 2, wherein the location/heading recognition unit uses a combination of a Particle Filter (PF) with an Extended Kalman Filter (EKF) to compare each projected image with the GIS map.

6. The apparatus of claim 2, wherein the location/heading recognition unit uses a linear Kalman filter upon estimating the dynamic behavioral state of the corresponding vehicle.

7. The apparatus of claim 1, wherein the wireless communication unit additionally receives path information including a recommended path from the in-housing complex management device, and transmits the path information to the location/heading recognition unit.

8. The apparatus of claim 7, wherein the recommended path includes a path along which the vehicle arrives at a destination, and a path along which the vehicle departs from the housing complex.

9. The apparatus of claim 1, wherein the wireless communication unit additionally receives auxiliary information with respect to interference with safe driving of the autonomous driving vehicle from the in-housing complex management device, and transmits the auxiliary information to the location/heading recognition unit.

10. The apparatus of claim 1, wherein the image sensor includes a plurality of sensors.

11. A system for providing location and heading information of an autonomous driving vehicle on a road within a housing complex, comprising:

an in-housing complex management device including an environment monitoring sensor for monitoring entry of an approved vehicle into a housing complex, and a management server for, as the entry of the approved vehicle into the housing complex is monitored by the environment monitoring sensor, generating a GIS map of inside of the housing complex and transmitting the GIS map to an in-autonomous driving vehicle device in a wireless manner; and

the in-autonomous driving vehicle device including an image sensor for detecting images of surroundings depending on motion of the autonomous driving vehicle, and a location/heading recognition unit for recognizing location and heading of the autonomous driving vehicle, based on the GIS map of the inside of the housing complex transmitted in a wireless manner from the in-housing complex management device and image information transmitted from the image sensor.

12. A method for providing location and heading information of an autonomous driving vehicle on a road within a housing complex, comprising:

detecting, by an image sensor installed on an autonomous driving vehicle, images of surroundings depending on motion of the autonomous driving vehicle;

receiving, by a wireless communication unit installed on the autonomous driving vehicle, a Geographic Information System (GIS) map of inside of the housing complex transmitted from an in-housing complex management device in a wireless manner; and

recognizing, by a location/heading recognition unit installed on the autonomous driving vehicle, location and heading of the autonomous driving vehicle based on image information about the detected images and the received GIS map of the inside of the housing complex.

13. The method of claim 12, wherein recognizing the location and heading of the autonomous driving vehicle comprises:

estimating ego-motion from stereoscopic images acquired by the image sensor, using a multi-view image processing technique;

extracting markers on the road by converting the images collected by the image sensor into images projected onto a ground surface of the road, and estimating a current location and attitude by obtaining a probability of matching the GIS map; and

estimating a dynamic behavioral state of the vehicle based on the estimated ego-motion and the estimated current location and attitude and then outputting predicted results of a state of the vehicle.

14. The method of claim 13, wherein estimating the ego-motion comprises:

defining three-dimensional (3D) coordinates of feature points detected from an image initially acquired by the image sensor as initial values of a point cloud;

tracking feature points acquired from a previous key frame on an image plane whenever each image is input from the image sensor, obtaining relative coordinates at a current time based on a key frame using a Perspective-Three-Point (P3P) technique, and calculating the ego-motion from relative coordinates of the previous key frame and relative coordinates at the current time; and

when movement of the autonomous driving vehicle above a predetermined amount occurs, and generation of a new key frame is required, performing bundle adjustment in a separate thread and then updating the point cloud.

15. The method of claim 13, wherein estimating the current location and attitude is configured to use a combination of a Particle Filter (PF) with an Extended Kalman Filter (EKF) to compare each projected image with the GIS map.

16. The method of claim 13, wherein estimating the current location and attitude comprises:

predicting posterior probabilities of a location and an attitude at which a quantity of ego-motion of the corresponding vehicle has been obtained;

obtaining weights of respective particles obtained via the prediction, and then updating the posterior probabilities; and

estimating the location and attitude of the vehicle based on predicted and updated results.

17. The method of claim 13, wherein outputting the predicted results is configured to use a linear Kalman filter upon estimating the dynamic behavioral state of the corresponding vehicle.

18. The method of claim 12, wherein receiving the GIS map comprises additionally receiving path information including a recommended path from the in-housing complex management device.

19. The method of claim 18, wherein the recommended path includes a path along which the vehicle arrives at a destination, and a path along which the vehicle departs from the housing complex.

20. The method of claim 12, wherein receiving the GIS map comprises additionally receiving auxiliary information with respect to interference with safe driving of the autonomous driving vehicle from the in-housing complex management device.