System and method to avoid obstacles in an autonomous unmanned maritime vehicle

1 . A system for detecting and avoiding an obstacle of an autonomous unmanned maritime vehicle (UMV) traveling in an initial direction, the system including:

at least one image sensor, the at least one image sensor configured to generate a plurality of video images;

a storage element configured to receive and store information associated with a plurality of predetermined object classes; and

a processing unit coupled to the at least one sensor and the storage element, the processing unit configured to:

receive at least one video image from the at least one image sensor, the video image containing at least one object;

determine if the at least one object can be associated with an object class from the plurality of predetermined object classes that has been identified as an obstacle;

access a mean height value for the object class if it is determined that the at least one object can be associated with an object class from the plurality of predetermined object classes that has been identified as an obstacle;

determine a distance between the UMV and the at least one object based on a height of the at least one object as displayed in the video image and the mean height for the object class; and

automatically adjust the navigational control of the autonomous UMV to travel in an adjusted direction.

2 . The system of claim 1 , further comprising a navigational control unit coupled to the processing unit, wherein the processing unit is further configured to;

generate at least one control signal based on the determined distance between the UMV and the at least one object; and

provide the at least one control signal to the navigational control unit in order to generate a navigational response in the UMV to the at least one object.

3 . The system of claim 2 wherein the navigational response is an evasive maneuver.

4 . The system of claim 1 , further comprising at least one sound navigation ranging (SONAR) sensor coupled to the processing unit, wherein the processing unit is further configured to:

receive data from the at least one SONAR sensor, the data including an indication of the presence of an object in the vicinity of the UMV;

calculate a distance between the UMV and the object in the vicinity based on the data received from the at least one SONAR sensor;

compare the calculated distance between the UMV and the object in the vicinity based on the data received from the at least one SONAR sensor to the determined distance between the UMV and the at least one object based on the height of the at least one object as displayed in the video image and the mean height for the object class; and

automatically adjust the navigational control of the autonomous UMV to travel in an adjusted direction based on the comparison.

5 . The system of claim 1 wherein the distance includes a distance in more than one dimension.

6 . The system of claim 1 , wherein the processing unit is further configured to access information about the image sensor from the storage unit, the information used to determine the height of the at least one object.

7 . The system of claim 6 , wherein the information about the image sensor includes at least one of position of the image sensor, field of view of the image sensor, focal length of the sensor, and pixel count of the image sensor.

8 . The system of claim 1 , wherein determining the distance between the UMV and the at least one object includes calculating the distance as a ratio of the mean height of the object class and the height, in vertical pixels, of the at least one object as displayed in the video image.

9 . The system of claim 8 , wherein the distance is calculated using the following formula: distance=(f*real height*image height)/(object height*sensor height); where: distance=distance from object to image sensor f=the focal distance of the camera, real height=estimate of the object height based on mean height of object class, image height=height of total image in pixels object height=number of vertical pixels the object occupies in the video image, and sensor height=the camera sensor height.

10 . The system of claim 1 , wherein the determination of whether the object falls within an object class is processed using a machine learning algorithm.

11 . The system of claim 1 , wherein the processing unit is further configured to classify the at least one object to generate a new object class if it is determined that the at least one object cannot be associated with an object class from the plurality of predetermined object classes.

12 . The system of claim 11 , wherein the processing unit is further configured to store the new object class in the storage element.

13 . The system of claim 11 , wherein the classification of the at least one object is processed using a machine learning algorithm.

14 . The system of claim 13 , wherein the machine learning algorithm is a you only look once (YOLO) algorithm.

15 . A method for detecting and avoiding an obstacle of an autonomous unmanned maritime vehicle (UMV) traveling in an initial direction, the method comprising:

receiving, a video image from at least one image sensor, the video image containing at least one object;

determining if the at least one object can be associated with an object class from the plurality of predetermined object classes that bas been identified as an obstacle;

accessing a mean height value for the object class if it is determined that the at least one object can be associated with an object class from the plurality of predetermined object classes that has been identified as an obstacle;

determining a distance between the UMV and the at least one object based on a height of the at least one object as displayed in the video image and the mean height for the object class; and

automatically adjusting the navigational control of the autonomous UMV to travel in an adjusted direction.

16 . The method of claim 15 , further comprising:

generating at least one control signal based on the determined distance between the UMV and the at least one object; and

providing the at least one control signal to a navigational control unit in order to generate a navigational response in the UMV to the at least one object.

17 . The method of claim 15 , further comprising:

receiving data from the at least one SONAR sensor, the data including an indication of the presence of an object in the vicinity of the UMV;

calculating a distance between the UMV and the object in the vicinity based on the data received from the at least one SONAR sensor;

comparing the calculated distance between the UMV and the object in the vicinity based on the data received from the at least one SONAR sensor to the determined distance between the UMV and the at least one object based on the height of the at least one object as displayed in the video image and the mean height for the object class; and

automatically adjusting the navigational control of the autonomous UMV to travel in an adjusted direction based on the comparison.

18 . The method of claim 15 , wherein determining the distance between the UMV and the at least one object includes calculating the distance as a ratio of the mean height of the object class and the height, in vertical pixels, of the at least one object as displayed in the video image.

19 . The method of claim 15 , further comprising classifying the at least one object to generate a new object class if it is determined that that the at least one object cannot be associated with an object class from the plurality of predetermined object classes.

20 . A method for detecting and avoiding an obstacle of an autonomous unmanned maritime vehicle (UMV) traveling in an initial direction, the method comprising:

receiving, a video image from at least one image sensor, the video image containing at least one object;

determining if the at least one object can be associated with an object class from the plurality of predetermined object classes;

accessing a mean height value for the object class if it is determined that the at least one object can be associated with an object class from the plurality of predetermined object classes;

determining a distance between the UMV and the at least one object based on a height of the at least one object as displayed in the video image and the mean height for the object class;

receiving data from the at least one SONAR sensor, the data including an indication of the presence of an object in the vicinity of the UMV;

calculating a distance between the UMV and the object in the vicinity based on the data received from the at least one SONAR sensor;

comparing the calculated distance between the UMV and the object in the vicinity based on the data received from the at least one SONAR sensor to the determined distance between the UMV and the at least one object based on a height of the at least one object as displayed in the video image and the mean height for the object class; and

automatically adjust the navigational control of the autonomous UMV to travel in an adjusted direction based on the comparison.