Back to results
US 12,592,151 B2Granted
System and method for multi-image-based vessel proximity situation recognition support
AssigneeKorea Institute of Ocean Science & Technology
InventorsNamSun Son, ChunSeon Pyo, HanSol Park
Life of this patent
Estimating the term…
Abstract
A system and method for multi-image-based vessel proximity situation recognition support is proposed. The system may include an unmanned surface vehicle (USV) configured to detect and track surrounding objects by monitoring surroundings using surrounding images and navigation sensors. The system may also include a remote navigation controller configured to support proximity situation recognition of the unmanned surface vehicle according to detection of the surrounding objects, wherein the unmanned surface vehicle may include an image acquisition processor, a navigation sensor, and a detector.
Claims5
A system for multi-image-based vessel proximity situation information support, the system comprising:
- an unmanned surface vehicle (USV) configured to detect and track surrounding objects by monitoring surroundings using surrounding images and navigation sensors;
- and
- a remote navigation controller configured to support proximity situation information of the unmanned surface vehicle according to detection of the surrounding objects,
- wherein the unmanned surface vehicle comprises:
- an image acquisition processor configured to acquire multiple images showing the surroundings of the unmanned surface vehicle to detect objects around the unmanned surface vehicle through image analysis;
- a navigation sensor configured to acquire current navigation information of the unmanned surface vehicle and information on obstacles around the unmanned surface vehicle in real time;
- and
- a detector configured to monitor the surroundings of the unmanned surface vehicle by using the multiple images, current navigation information, and information on obstacles and, when an object close to the unmanned surface vehicle within a preset distance is detected as a result of monitoring, track the object detected, the detector further configured to detect and track one or more additional moving objects within the preset distance,
- wherein the remote navigation controller comprises:
- an estimation processor configured to estimate a collision risk between the unmanned surface vehicle and the object detected by using the multiple images, current navigation information, and information on obstacles;
- and
- a situation information processor configured to display the collision risk estimated, along with the object detected, on an electronic navigational chart for each detected object and to output the collision risk in order to support proximity situation information,
- wherein the estimation processor is configured to:
- determine whether the object detected is located on an expected path of the unmanned surface vehicle by using the current navigation information,
- based on determining that the object detected is located on the expected path, calculate the collision risk between the unmanned surface vehicle and the object detected, by using fuzzy inference, and
- based on determining that the detected object is identified as a moving object, calculate a time when the detected moving object passes through the expected path of the unmanned surface vehicle, and calculate the collision risk between the unmanned surface vehicle and the detected moving object at the time of passing through the calculated predicted path by the detected moving object, by using fuzzy inference, and
- wherein the remote navigation controller is configured to:
- determine that the distance between the unmanned surface vehicle and the detected moving object is less than a distance between the unmanned surface vehicle and any of the detected one or more additional moving objects,
- calculate collision risks between the unmanned surface vehicle and the detected one or more additional moving objects,
- determine that the collision risk between the unmanned surface vehicle and the detected moving object is higher than any of the collision risks between the unmanned surface vehicle and the detected one or more additional moving objects,
- determine the detected moving object to have a high risk of collision,
- display the detected moving object with a warning sign indicating a collision danger,
- display the calculated collision risk on an augmented reality or virtual reality screen based on the multiple images showing the surroundings of the unmanned surface vehicle to support a remote operator's awareness,
- determine that the detected object is located on the expected path of the unmanned surface vehicle based on the current navigation information including all of the location, speed, direction, and posture of the unmanned surface vehicle, and
- transmit control commands and control information to the unmanned surface vehicle to adjust at least one of a speed or a direction of the unmanned surface vehicle to avoid a collision between the unmanned surface vehicle and the detected moving object such that the unmanned surface vehicle is controlled by the control commands and control information received from the remote navigation controller.
22 more elements
A method for multi-image-based vessel proximity situation information support performed in a system including an unmanned surface vehicle and a remote navigation control device, the method comprising:
- acquiring, by the unmanned surface vehicle, multiple images of surroundings to detect surrounding objects through image analysis;
- obtaining, by the unmanned surface vehicle, current navigation information and information on surrounding obstacles in real time;
- monitoring, by the unmanned surface vehicle, the surroundings using the multiple images, current navigation information, and information on surrounding obstacles and, when an object close to a surrounding area within a preset distance is detected as a result of monitoring, tracking the object detected, the monitoring comprising detecting and tracking one or more additional moving objects within the preset distance;
- supporting, by the remote navigation control device, proximity situation information of the unmanned surface vehicle according to detection of the object;
- estimating, by an estimation processor of the remote navigation control device, a collision risk between the unmanned surface vehicle and the object detected by using the multiple images, current navigation information, and information on obstacles;
- displaying, by a situation information processor of the remote navigation control device, the collision risk estimated, along with the object detected, on an electronic navigational chart for each detected object and to output the collision risk in order to support proximity situation information;
- determining, by the estimation processor, whether the object detected is located on an expected path of the unmanned surface vehicle by using the current navigation information;
- based on determining that the object detected is located on the expected path, calculating, by the estimation processor, the collision risk between the unmanned surface vehicle and the object detected, by using fuzzy inference;
- based on determining that the detected object is identified as a moving object, calculating, by the estimation processor, a time when the detected moving object passes through the expected path of the unmanned surface vehicle, and calculating the collision risk between the unmanned surface vehicle and the detected moving object at the time of passing through the calculated predicted path by the detected moving object, by using fuzzy inference;
- determining, by the remote navigation control device, that the distance between the unmanned surface vehicle and the detected moving object is less than a distance between the unmanned surface vehicle and any of the detected one or more additional moving objects;
- calculating, by the remote navigation control device, collision risks between the unmanned surface vehicle and the detected one or more additional moving objects;
- determining, by the remote navigation control device, that the collision risk between the unmanned surface vehicle and the detected moving object is higher than any of the collision risks between the unmanned surface vehicle and the detected one or more additional moving objects;
- determining, by the remote navigation control device, the detected moving object to have a high risk of collision;
- displaying, by the remote navigation control device, the detected moving object with a warning sign indicating a collision danger,
- displaying the calculated collision risk on an augmented reality or virtual reality screen based on the multiple images showing the surroundings of the unmanned surface vehicle to support a remote operator's awareness;
- determining that the detected object is located on the expected path of the unmanned surface vehicle based on the current navigation information including all of the location, speed, direction, and posture of the unmanned surface vehicle;
- and
- transmitting, by the remote navigation control device, control commands and control information to the unmanned surface vehicle to adjust at least one of a speed or a direction of the unmanned surface vehicle to avoid a collision between the unmanned surface vehicle and the detected moving object such that the unmanned surface vehicle is controlled by the control commands and control information received from the remote navigation control device.
15 more elements
Dependencies are inferred from the claim text.