Unmanned ship berthing method and system

Abstract

The invention discloses an unmanned ship berthing method and system, and relates to the technical field of fishery breeding, and the method comprises the steps: projecting a three-dimensional point cloud of a net cage in a front view field of a laser radar carried by an unmanned ship to a two-dimensional plane, and dividing a coordinate system into an upper region, a lower region and a region near the axis; the scanning frequency of the laser radar is set, the unmanned ship is driven to berth towards the net cage, and the mean value of the abscissa values of all the three-dimensional point clouds in each area is continuously counted; comparing the mean value in the upper area with the mean value in the lower area, and controlling the steering of the unmanned ship; determining a difference value between a mean value in the upper region and a mean value in the lower region, and determining a relative posture of the unmanned ship and the net cage edge according to a ratio of the difference value to the mean value in the region near the axis; and when the relative attitude and the mean value in the area near the axis representing the position of the unmanned ship both meet the required conditions, the unmanned ship and the net cage are mechanically locked. According to the invention, automatic positioning and attitude determination of the unmanned ship can be realized.

Description

Technical Field

[0001] The present invention relates to the technical field of fishery breeding, and in particular to a berthing method and system for an unmanned boat. Background Art

[0002] Cage farming is an efficient aquaculture method. For cage farming, after planning site selection, cage setting, and stocking fry, daily management is the most time-consuming and labor-intensive key step, which mainly includes: feed feeding and water quality monitoring. The existing invention discloses an aquaculture feed feeding method and system based on an unmanned ship. First, it continuously analyzes whether there are target aquatic products in the image information, and when there are target aquatic products in the image information, it obtains the aggregation level information of the target aquatic products according to the image information analysis, and then obtains the feeding amount information according to the aggregation level information and the feeding plan. This method can effectively solve the problem of feed waste caused by the absence of aquatic animals or too few aquatic animals near the preset delivery point, and the problem of insufficient feed and affected aquaculture effect of aquatic animals due to the excessive number of aquatic animals near the preset delivery point, thereby effectively improving the accuracy of feed delivery.

[0003] However, existing technologies cannot achieve automatic positioning and attitude determination of the unmanned boat according to the position of the cage. Summary of the invention

[0004] The embodiment of the present invention provides a berthing method and system for an unmanned boat, which solves the problem in the prior art that the unmanned boat cannot be automatically positioned and determined according to the position of the cage.

[0005] An embodiment of the present invention provides a method for berthing an unmanned boat, comprising the following steps: Obtain the three-dimensional point cloud of the cage in the front field of view of the laser radar carried by the unmanned boat; establish a coordinate system on the two-dimensional plane with the position of the sensor on the unmanned boat as the coordinate origin, regard the direction of the bow as the X-axis, and the direction perpendicular to the X-axis as the Y-axis; project the three-dimensional point cloud of the cage to the coordinate system; Divide the coordinate system into an upper area, a lower area and an area near the axis; Set the scanning frequency of the laser radar and drive the unmanned boat to berth to the cage, and continuously count the mean of the horizontal coordinate values ​​of all three-dimensional point clouds in each area; According to the size of the mean value in the upper area and the mean value in the lower area, the unmanned boat is controlled to turn to the side where the center point of the cage is located during navigation; the difference between the mean value in the upper area and the mean value in the lower area is obtained, and the relative posture of the unmanned boat and the edge of the cage during navigation is determined according to the ratio of the difference to the mean value in the area near the axis; When the mean value in the area near the axis is less than the set distance threshold, and when the ratio is less than the set ratio threshold, the unmanned boat and the cage are controlled to be mechanically locked.

[0006] Further, the steps of dividing the coordinate system into an upper region, a lower region, and a region near the axis specifically include: When the ordinate value of each point in the two-dimensional plane is greater than 1, the region composed of all points that meet the condition is divided into the upper region; when the ordinate value of each point in the two-dimensional plane is less than -1, the region composed of all points that meet the condition is divided into the lower region; when the ordinate value of each point in the two-dimensional plane is less than or equal to 1 and greater than or equal to -1, the region composed of all points that meet the condition is divided into the region near the axis.

[0007] Further, the steps of controlling the unmanned boat to turn towards the side where the center point of the fish cage is located during navigation specifically include: When the mean value in the upper region is greater than the mean value in the lower region, control the unmanned boat to turn in the direction of the upper region; when the mean value in the upper region is less than the mean value in the lower region, control the unmanned boat to turn in the direction of the lower region; when the mean value in the upper region is equal to the mean value in the lower region, control the unmanned boat to go straight.

[0008] Further, the steps of determining the relative attitude of the unmanned boat with respect to the edge of the fish cage during navigation specifically include: Denote the mean value in the upper region as and denote the mean value in the lower region as and denote the mean value in the region near the axis as ; the relative attitude of the unmanned boat with respect to the edge of the fish cage is expressed as When is 0, it means that the heading of the unmanned boat points to the center point of the fish cage.

[0009] Further, the steps of controlling the mechanical locking of the unmanned boat with the fish cage specifically include: Set an error threshold and take the distance between the installation position of the sensor on the unmanned boat and the fixed teeth on the boom of the unmanned boat as the set minimum threshold ; the smaller the mean value in the region near the axis, the closer the position of the unmanned boat is to the fish cage; when , control the boom of the unmanned boat to press down so that the fixed teeth on the boom are mechanically locked with the fish cage.

[0010] Further, the steps of projecting the three-dimensional point cloud of the fish cage onto the coordinate system specifically include: Project the three-dimensional point cloud Pi(x, y, z) of the fish cage within a 90-degree forward field of view of the lidar onto the coordinate system to obtain the corresponding point B(x, y); assign 1 to B(x, y) and assign 0 to the position on the sea surface where there is no radar echo.

[0011] An embodiment of the present invention provides a mooring system for an unmanned boat, including: A point cloud projection module, configured to obtain the three-dimensional point cloud of the net cage within the front field of view of the lidar carried by the unmanned boat; establish a coordinate system in a two-dimensional plane with the position where the sensor on the unmanned boat is located as the coordinate origin, regard the direction where the bow is located as the X-axis, and the direction perpendicular to the X-axis as the Y-axis; project the three-dimensional point cloud of the net cage onto the coordinate system. A region division module, configured to divide the coordinate system into an upper region, a lower region, and a region near the axis. A pose locking module, configured to set the scanning frequency of the lidar and drive the unmanned boat to moor towards the net cage, continuously count the mean value of the abscissa values of all the three-dimensional point clouds in each region; control the unmanned boat to turn towards the side where the center point of the net cage is located during navigation according to the magnitudes of the mean value in the upper region and the mean value in the lower region; obtain the difference between the mean value in the upper region and the mean value in the lower region, and determine the relative pose of the unmanned boat to the edge of the net cage during navigation according to the ratio of the difference to the mean value in the region near the axis; when the mean value in the region near the axis is less than the set distance threshold and when the ratio is less than the set ratio threshold, control the mechanical locking of the unmanned boat and the net cage.

[0012] An embodiment of the present invention provides a mooring method and system for an unmanned boat. Compared with the prior art, its beneficial effects are as follows: The coordinate system is divided into an upper region, a lower region, and a region near the axis; during the process of the unmanned boat approaching the net cage, continuously count the mean value of the abscissa values of all the three-dimensional point clouds in each region; the turning of the unmanned boat is determined by the magnitudes of the mean value in the upper region and the mean value in the lower region, and the relative pose of the unmanned boat to the edge of the net cage during navigation is determined by the mean value in the upper region, the mean value in the lower region, and the mean value in the region near the axis. The mean value of the abscissa values of all the three-dimensional point clouds in each region changes with the continuously changing distance between the unmanned boat and the net cage, and can continuously adjust the pose and position of the unmanned boat according to the mean value in each region during navigation until the pose of the unmanned boat relative to the net cage and the distance between the unmanned boat and the net cage meet the requirements, and control the mechanical locking of the unmanned boat and the net cage to complete the positioning and pose determination of the unmanned boat. Description of the Drawings

[0013] Figure 1 It is a schematic diagram of cage aquaculture provided by an embodiment of the present invention; Figure 2 It is a binarized two-dimensional plan view of the mooring method of the unmanned boat provided by an embodiment of the present invention; Figure 3 It is a flowchart of the mooring method of the unmanned boat provided by an embodiment of the present invention. Detailed Embodiments

[0014] To make the above objects, features, and advantages of the present invention more apparent and understandable, the following describes the specific embodiments of the present invention in detail with reference to the accompanying drawings. Many specific details are set forth in the following description to facilitate a full understanding of the present invention. However, the present invention can be implemented in many other ways different from those described herein, and those skilled in the art can make similar improvements without departing from the spirit of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0015] An embodiment of the present invention provides a berthing method for an unmanned boat, including the following steps: Step 1: Obtain the three-dimensional point cloud of the net cage within the forward field of view of the lidar carried by the unmanned boat; establish a coordinate system in a two-dimensional plane with the position of the sensor on the unmanned boat as the coordinate origin, regard the direction where the bow is located as the X-axis, and the direction perpendicular to the X-axis as the Y-axis; project the three-dimensional point cloud of the net cage onto the coordinate system.

[0016] Step 2: Divide the coordinate system into an upper region, a lower region, and a region near the axis.

[0017] Step 3: Set the scanning frequency of the lidar and drive the unmanned boat to berth towards the net cage, continuously count the mean value of the abscissa values of all three-dimensional point clouds in each region; according to the magnitudes of the mean values in the upper region and the lower region, control the unmanned boat to turn towards the side where the center point of the net cage is located during navigation; obtain the difference between the mean value in the upper region and the mean value in the lower region, and determine the relative attitude of the unmanned boat to the edge of the net cage during navigation according to the ratio of the difference to the mean value in the region near the axis; when the mean value in the region near the axis is less than the set distance threshold and when the ratio is less than the set ratio threshold, control the mechanical locking of the unmanned boat and the net cage.

[0018] 1. The overall technical solution of the unmanned boat for cage aquaculture consists of three parts: the net cage, the unmanned boat, and the base station, as Figure 1 shown.

[0019] The core lies in how the unmanned boat conducts water quality detection and feeding on the net cage. The overall working steps are as follows: (1) The twin-hull aquaculture unmanned boat departs from the base station, conducts autonomous path planning and navigation according to the position information of the net cage. The position information of the net cage can be measured manually in advance or sent in real time by installing a GPS on the net cage.

[0020] (2) After the unmanned boat approaches the target net cage, it senses the net cage frame through the on-board lidar, and automatically controls the heading of the unmanned boat through the net cage pose calculation method, so that the unmanned boat sails perpendicular to the edge of the net cage, and the distance gradually decreases. When the distance between the unmanned boat and the net cage frame meets the requirements, send a command to control the boom motor.

[0021] (3)When the boom touches the railing of the cage frame, it is locked to achieve the pose locking of the unmanned boat and the cage, avoiding the random drift of the unmanned boat under the interference of wind, waves and currents, and creating conditions for subsequent work.

[0022] (4)After completing the water quality detection and feed feeding tasks of this cage, the nacelle is retracted and the boom is lifted. The unmanned boat and the cage are unlocked, and the unmanned boat patrols other cages according to the mission plan (repeat step 2).

[0023] (5)When the conditions are met (mission completed / energy about to run out / feed used up), the unmanned boat returns to the base station by itself to replenish feed, energy, and upload all water quality detection data of the cages.

[0024] Among them: This solution is applicable to traditional floating gravity cages. The cage frame floats on the water surface, the netting is hung by the frame, and a certain cage shape and volume are maintained by the buoyancy of the box body itself and the weight of the netting.

[0025] 2. The unmanned boat adopts a catamaran structure to ensure sufficient anti-wind and wave characteristics and sufficient equipment installation space.

[0026] 3. When the boom presses down at an appropriate distance, it will be locked on the cage frame to maintain a stable attitude and spacing between the unmanned boat and the cage, avoiding the influence of wind, waves and currents during subsequent work. After the boom is lifted, the unmanned boat can be unlocked from the cage and resume free navigation. The boom movement is controlled by the on-board host: when the cage enters the effective range, the boom lifting motor is controlled to press down, and it rises after completing the water quality detection / feeding work.

[0027] 4. The environmental perception module includes water surface sensors such as lidar and cameras. The data is input into the on-board host for real-time calculation. During navigation, the unmanned boat can achieve automatic driving and obstacle avoidance. When approaching the cage, it relies on the cage pose perception algorithm to determine whether the boom pressing condition is met.

[0028] 5. The flow of the cage pose perception algorithm is as Figure 3 shown, and the specific steps are as follows: (1)The lidar obtains the point cloud of the front view 90 degrees (45 degrees on each side) field of view: Pi(x, y, z).

[0029] (2)The three-dimensional point cloud is projected onto the XY plane to form a top view B(x, y). Generally, there is no radar echo on the sea surface, which can be assigned a value of 0, and there must be radar echoes for guardrails, etc., which are set to 1. When the unmanned boat approaches the cage, the 2-value image of the lidar point cloud projected onto the top view is as Figure 2 shown.

[0030] (3) Establish a carrier coordinate system with the sensor position as the origin, the bow direction as the X-axis, and the positive direction of the Y-axis as the right side of the boat. Respectively, count the average X-distance of the valid points in the three regions above the Y-axis, below the Y-axis, and near the axis:

[0031] ) ) )

[0032] (4) During the process of the unmanned boat approaching the cage, continuously (10Hz) calculate the above 3 relative distance values, and control the heading of the unmanned boat according to the following method: If then turn right; if then turn left; if then go straight.

[0033] (5) If , then the condition for pressing down the boom is satisfied.

[0034] In the formula: represents the relative attitude between the longitudinal axis of the unmanned boat and the edge of the cage, which is 0 when they are perpendicular to each other (set a small threshold to tolerate errors), represents the distance from the lidar to the cage, and the minimum threshold can be set by measuring the distance from the sensor installation position to the fixed teeth of the boom .

[0035] The effects of the present invention are as follows: (1) Relying only on the sensors and feeding devices carried by the unmanned boat itself, an unmanned boat can realize autonomous water quality detection and feed feeding for multiple traditional cages.

[0036] (2) Through the boom form, when the unmanned boat is in open water, it can detect and feed the internal area of the cage.

[0037] (3) By the method of pressing down and lifting the boom, the pose locking and unlocking between the unmanned boat and the cage can be realized under windy and wavy conditions.

[0038] (4) Through the lidar point cloud, the relative pose between the unmanned boat and the cage is obtained to ensure that the boom can correctly contact the cage frame.

[0039] (5) The boom integrates a water quality detection pod, and the depth of the pod is controllable, realizing multi-point three-dimensional measurement of the water quality of the cage.

[0040] (6) The boom integrates a feed feeding port, and the feed is put into the cage by the wind-driven method.

[0041] An embodiment of the present invention provides a mooring system for an unmanned boat, including: A point cloud projection module, configured to obtain the three-dimensional point cloud of the net cage within the front view field of the lidar carried by the unmanned boat; establish a coordinate system in a two-dimensional plane with the position of the sensor on the unmanned boat as the coordinate origin, regard the direction where the bow is located as the X-axis, and regard the direction perpendicular to the X-axis as the Y-axis; project the three-dimensional point cloud of the net cage onto the coordinate system.

[0042] A region division module, configured to divide the coordinate system into an upper region, a lower region, and a region near the axis.

[0043] A pose locking module, configured to set the scanning frequency of the lidar and drive the unmanned boat to moor to the net cage, continuously count the mean value of the abscissa values of all the three-dimensional point clouds in each region; according to the magnitudes of the mean values in the upper region and the lower region, control the unmanned boat to turn towards the side where the center point of the net cage is located during navigation; obtain the difference between the mean value in the upper region and the mean value in the lower region, and determine the relative pose between the unmanned boat and the edge of the net cage according to the ratio of the difference to the mean value in the region near the axis; when the mean value in the region near the axis is less than the set distance threshold and when the ratio is less than the set ratio threshold, control the mechanical locking of the unmanned boat and the net cage.

[0044] A specific embodiment is as follows: An embodiment of the present invention discloses a mooring method for an unmanned boat, and the specific steps include: S1. Project the three-dimensional point cloud of the net cage within the front view field of the lidar carried by the unmanned boat onto a two-dimensional plane.

[0045] S2. Establish a coordinate system in a two-dimensional plane with the position of the sensor on the unmanned boat as the coordinate origin, regard the direction where the bow is located as the X-axis, and regard the direction perpendicular to the X-axis as the Y-axis. And divide the coordinate system into an upper region, a lower region, and a region near the axis.

[0046] S3. Drive the unmanned boat to moor to the net cage, and continuously count the mean value of the abscissa values of all the three-dimensional point clouds in each region.

[0047] S4. Compare the magnitudes of the mean value in the upper region and the mean value in the lower region, and control the turning of the unmanned boat according to the comparison result; determine the difference between the mean value in the upper region and the mean value in the lower region, and determine the relative pose between the unmanned boat and the edge of the net cage according to the ratio of the difference to the mean value in the region near the axis.

[0048] S5. When both the relative pose and the mean value in the region near the axis representing the position of the unmanned boat meet the required conditions, mechanically lock the unmanned boat and the net cage.

[0049] The above-described embodiments merely represent several implementation manners of the present invention. The description thereof is relatively specific and detailed, but it should not be construed as a limitation to the scope of the invention patent. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present invention, several modifications and improvements can still be made, and these all fall within the protection scope of the present invention. Therefore, the protection scope of the present invention patent shall be subject to the appended claims.

Claims

1. A method for berthing an unmanned boat, characterized in that: The following steps are involved: Obtain the three-dimensional point cloud of the net box in the front field of view of the laser radar carried by the unmanned boat; A coordinate system is established on a two-dimensional plane with the location of the sensor on the unmanned boat as the origin of the coordinates, the direction of the bow is regarded as the X-axis, and the direction perpendicular to the X-axis is regarded as the Y-axis; the three-dimensional point cloud of the cage is projected onto the coordinate system; Divide the coordinate system into an upper area, a lower area and an area near the axis; Set the scanning frequency of the laser radar and drive the unmanned boat to berth to the cage, and continuously count the mean of the horizontal coordinate values ​​of all three-dimensional point clouds in each area; According to the size of the mean value in the upper area and the mean value in the lower area, the unmanned boat is controlled to turn toward the side where the center point of the cage is located during navigation; Obtain the difference between the mean value in the upper area and the mean value in the lower area, and determine the relative posture of the unmanned boat and the edge of the cage during navigation according to the ratio of the difference to the mean value in the area near the axis; When the mean value in the area near the axis is less than the set distance threshold, and when the ratio is less than the set ratio threshold, the unmanned boat and the cage are controlled to be mechanically locked.

2. The berthing method of an unmanned boat according to claim 1, characterized in that: The specific steps of dividing the coordinate system into an upper area, a lower area and an area near the axis include: When the ordinate value of each point in the two-dimensional plane is greater than 1, the area consisting of all points that meet the conditions is divided into the upper area; When the ordinate value of each point in the two-dimensional plane is less than -1, the area consisting of all points that meet the conditions is divided into the lower area; When the ordinate value of each point in the two-dimensional plane is less than or equal to 1 and greater than or equal to -1, the area consisting of all points that meet the conditions is divided into the area near the axis.

3. The berthing method of an unmanned boat according to claim 1, characterized in that: The controlling of the unmanned boat to turn toward the side where the center point of the cage is located during navigation specifically comprises the following steps: When the mean value in the upper area is greater than the mean value in the lower area, the unmanned boat is controlled to rotate in the direction of the upper area; When the mean value in the upper area is smaller than the mean value in the lower area, the unmanned boat is controlled to rotate toward the direction of the lower area; When the mean value in the upper area is equal to the mean value in the lower area, the unmanned boat is controlled to move straight.

4. The berthing method of an unmanned boat according to claim 1, characterized in that: The specific steps of determining the relative posture of the unmanned boat and the edge of the cage during navigation include: The mean value in the upper region is recorded as , and the mean value in the lower region is recorded as , and the mean value in the area near the axis is recorded as ; The relative posture between the unmanned boat and the edge of the cage is expressed as ,when When it is 0, it means that the heading of the unmanned boat points to the center of the cage.

5. The berthing method of an unmanned boat as claimed in claim 4, characterized in that: The control of the unmanned boat and the cage mechanical locking specifically comprises the following steps: Setting the error threshold The minimum threshold is set as the distance between the installation position of the sensor on the unmanned ship and the fixed teeth on the boom of the unmanned ship. ; The mean value of the area near the axis The smaller it is, the closer the unmanned boat is to the cage; when When the crane arm of the unmanned boat is controlled to be pressed down, the fixed teeth on the crane arm are mechanically locked with the cage.

6. The berthing method of an unmanned boat according to claim 1, characterized in that: The three-dimensional point cloud of the cage is projected to the coordinate system, and the specific steps include: Project the three-dimensional point cloud Pi (x, y, z) of the box within the 90-degree field of view in front of the laser radar to the coordinate system, and obtain the corresponding point B (x, y); Assign B(x, y) a value of 1, and assign a value of 0 to the location on the sea surface where there is no radar echo.

7. A berthing system for an unmanned boat, characterized in that: include: Point cloud projection module, used to obtain the three-dimensional point cloud of the net box in the front field of view of the laser radar carried by the unmanned boat; A coordinate system is established on a two-dimensional plane with the location of the sensor on the unmanned boat as the origin of the coordinates, the direction of the bow is regarded as the X-axis, and the direction perpendicular to the X-axis is regarded as the Y-axis; the three-dimensional point cloud of the cage is projected onto the coordinate system; A region division module is used to divide the coordinate system into an upper region, a lower region and a region near the axis; The posture locking module is used to set the scanning frequency of the laser radar and drive the unmanned boat to berth to the cage, continuously counting the mean of the horizontal coordinate values ​​of all three-dimensional point clouds in each area; according to the size of the mean in the upper area and the mean in the lower area, the unmanned boat is controlled to turn to the side where the center point of the cage is located during navigation; Obtain the difference between the mean value in the upper area and the mean value in the lower area, and determine the relative posture of the unmanned boat and the edge of the cage during navigation according to the ratio of the difference to the mean value in the area near the axis; When the mean value in the area near the axis is less than the set distance threshold, and when the ratio is less than the set ratio threshold, the unmanned boat and the cage are controlled to be mechanically locked.

Images