CN116859948A - Autonomous navigation control method and system for unmanned ship for channel sweep based on target detection algorithm - Google Patents
Autonomous navigation control method and system for unmanned ship for channel sweep based on target detection algorithm Download PDFInfo
- Publication number
- CN116859948A CN116859948A CN202310983281.6A CN202310983281A CN116859948A CN 116859948 A CN116859948 A CN 116859948A CN 202310983281 A CN202310983281 A CN 202310983281A CN 116859948 A CN116859948 A CN 116859948A
- Authority
- CN
- China
- Prior art keywords
- ship
- obstacle
- navigation
- degrees
- azimuth angle
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 68
- 238000000034 method Methods 0.000 title claims abstract description 40
- 238000013135 deep learning Methods 0.000 claims abstract description 6
- 238000004458 analytical method Methods 0.000 claims abstract description 4
- 238000004891 communication Methods 0.000 claims description 25
- 238000012545 processing Methods 0.000 claims description 9
- 230000004888 barrier function Effects 0.000 claims description 3
- 238000003384 imaging method Methods 0.000 claims description 3
- 238000005259 measurement Methods 0.000 abstract description 3
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 16
- 230000000873 masking effect Effects 0.000 description 10
- 230000008569 process Effects 0.000 description 9
- 238000010586 diagram Methods 0.000 description 8
- 230000006870 function Effects 0.000 description 6
- 230000033001 locomotion Effects 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 4
- 238000000605 extraction Methods 0.000 description 4
- 238000009827 uniform distribution Methods 0.000 description 3
- 238000012544 monitoring process Methods 0.000 description 2
- 238000007781 pre-processing Methods 0.000 description 2
- 238000010408 sweeping Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 210000004556 brain Anatomy 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Landscapes
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
Abstract
The invention discloses an autonomous navigation control method and an autonomous navigation control system of a channel unmanned ship based on a target detection algorithm, wherein the method and the system are characterized in that channel surrounding environment data are acquired through a camera, the channel surrounding environment data are input into GPU embedded equipment carried by a ship, the type of a navigation obstacle in an image frame is identified by using a deployed deep learning target detection algorithm, and the position of the navigation obstacle in the image is recorded; the navigation obstacle is detected through the laser radar, the distance and azimuth information of the navigation obstacle are recorded, the combination analysis is carried out on the distance and azimuth information and the image recognition result, the attribute information of the navigation obstacle is obtained, and unmanned ship navigation preset obstacle avoidance measures are adopted according to the obtained attribute information of the navigation obstacle. According to the method provided by the invention, the specific type of the navigation-obstacle is identified by utilizing the high-performance GPU embedded terminal carried by the unmanned ship and the corresponding target detection algorithm, so that more accurate control is realized, and meanwhile, the laser radar is carried, so that the distance measurement of the navigation-obstacle is more accurate.
Description
Technical Field
The invention relates to the technical field of environment sensing and path planning, in particular to a channel sweep unmanned ship autonomous navigation control method and system based on a target detection algorithm.
Background
In the aspect of channel management and construction, the water depth measurement is carried out on each position in the channel mainly by means of engineering measuring ships driven by workers in person. Although a very small number of channel bureaus can measure the water depth in the channel by using a remote measuring and remote control technology to remotely control engineering survey vessels, the obstacle avoidance method adopted by the unmanned ship in the autonomous navigation process is very simple, and basically, an ultrasonic sensor is used for detecting the obstacle to avoid the obstacle. In addition, the route of the existing unmanned ship is input to the ship control end in advance in a preset mode, so that the unmanned ship is not flexible. Because the detection distance of the ultrasonic sensor is limited, and the specific type of the obstacle cannot be judged, different avoidance measures cannot be made for different navigation-impeding objects.
Therefore, an autonomous navigation control method and an autonomous navigation control system for a channel-scan unmanned ship based on a target detection algorithm are needed.
Disclosure of Invention
In view of the above, the invention aims to provide a method and a system for autonomous navigation control of a channel-sweeping unmanned ship based on a target detection algorithm.
In order to achieve the above purpose, the present invention provides the following technical solutions:
the invention provides a method for controlling autonomous navigation of a channel unmanned ship based on a target detection algorithm, which comprises the following steps:
acquiring channel surrounding environment data through a camera, inputting the channel surrounding environment data into a GPU embedded device carried by a ship, identifying the type of the navigation obstacle in an image frame by using a deployed deep learning target detection algorithm, and recording the position of the navigation obstacle in the image;
the navigation obstacle is detected through the laser radar, the distance and azimuth information of the navigation obstacle are recorded, the combination analysis is carried out on the distance and azimuth information and the image recognition result, the attribute information of the navigation obstacle is obtained, and unmanned ship navigation preset obstacle avoidance measures are adopted according to the obtained attribute information of the navigation obstacle.
Further, the attribute information of the navigation obstacle comprises the category of the navigation obstacle, the distance and the azimuth of the navigation obstacle and the unmanned ship.
Further, the preset obstacle avoidance measures are carried out according to the following steps:
if the obstruction is a navigation mark/buoy, this is done in the following way:
if the distance of the navigation obstacle is more than 10m, the azimuth angle of the navigation obstacle is within 15 degrees of the port; the left motor of the ship keeps unchanged speed, the speed of the right motor is reduced to 0, and the ship resumes direct voyage after the azimuth angle of the object in the way of voyage is increased to 30 degrees on the port;
If the distance of the navigation obstacle is more than 10 meters, the azimuth angle of the navigation obstacle is within 15 degrees of the starboard; the right motor of the ship keeps unchanged speed, the speed of the left motor is reduced to 0, and the ship resumes direct voyage after the azimuth angle of the obstacle is increased to 30 degrees on the starboard;
if the distance of the navigation obstacle is less than 10 meters, the azimuth angle of the navigation obstacle is within 15 degrees of the port; executing the speed increase of the left motor of the ship by 20%, and reducing the speed of the right motor to 0 until the azimuth angle of the navigation-impaired object is increased to 30 degrees on the port side, and then restoring the direct navigation;
if the distance of the navigation obstacle is less than 10 meters, the azimuth angle of the navigation obstacle is within 15 degrees of the starboard; the speed of the right motor of the ship is increased by 20 percent, the speed of the left motor is reduced to 0, and the ship resumes straight voyage after the azimuth angle of the object in the way is increased to 30 degrees on the starboard.
Further, the preset obstacle avoidance measures are carried out according to the following steps:
if the obstruction is a lightboat, this is done in the following way:
if the distance of the navigation obstacle is more than 10m, the azimuth angle of the navigation obstacle is within 15 degrees of the port; the left motor of the ship keeps increasing the speed by 10 percent, the speed of the right motor is reduced to 0, and the ship resumes direct voyage after the azimuth angle of the object in the way of voyage is increased to 30 degrees on the port;
if the distance of the navigation obstacle is more than 10 meters, the azimuth angle of the navigation obstacle is within 15 degrees of the starboard; the right motor of the ship keeps increasing the speed by 10 percent, the speed of the left motor is reduced to 0, and the ship resumes direct voyage after the azimuth angle of the object in the way of voyage is increased to 30 degrees on the starboard;
If the distance of the navigation obstacle is less than 10 meters, the azimuth angle of the navigation obstacle is within 15 degrees of the port; the left motor of the ship keeps increasing speed by 30%, the speed of the right motor is reduced to 0, and the ship resumes direct voyage after the azimuth angle of the object in the way is increased to 30 degrees on the port;
if the distance of the navigation obstacle is less than 10 meters, the azimuth angle of the navigation obstacle is within 15 degrees of the starboard; the right motor of the ship is kept to be increased by 30 percent, the speed of the left motor is reduced to 0, and the ship resumes straight voyage until the azimuth angle of the object in the way is increased to 30 degrees on the starboard.
Further, the preset obstacle avoidance measures are carried out according to the following steps:
if the obstacle is a cargo ship/mail ship, this is done in the following way:
if the distance of the obstacle is more than 30m, the azimuth angle of the obstacle is within 15 degrees of the port (namely, the azimuth angle of the obstacle is more than 0 degree of the port and less than 15 degrees of the port, and the situation is a front meeting state or a overtaking state); the left motor of the ship keeps increasing the speed by 20 percent, the speed of the right motor is reduced to 0, and the direct voyage is restored after the azimuth angle of the sailing obstacle becomes larger than 30 degrees of the port;
if the distance of the obstacle is more than 30 meters, the azimuth angle of the obstacle is within 15 degrees of starboard (namely, the azimuth angle of the obstacle is more than 0 degree of starboard and less than 15 degrees of starboard, and the situation is a front meeting state or a overtaking state); the right motor of the ship is kept at a speed increasing of 20%, the speed of the left motor is reduced to 0, and the direct voyage is restored after the azimuth angle of the sailing obstacle becomes larger than 30 degrees on the starboard;
If the space of the obstacle is less than 30 meters, the space angle of the obstacle is within 15 degrees of the port (namely, the space angle of the obstacle is greater than 0 degree of the port and less than 15 degrees of the port, and the situation is a front meeting state or a overtaking state); then the speed of the left motor of the ship is increased by 40%, the speed of the right motor is reduced to 0, and the ship resumes direct voyage after the azimuth angle of the sailing obstacle becomes larger than 30 degrees of the port;
if the distance of the obstacle is less than 30 meters, the azimuth angle of the obstacle is within 15 degrees of starboard (namely, the azimuth angle of the obstacle is greater than 0 degree of starboard and less than 15 degrees of starboard, and the situation is a front meeting state or a overtaking state); the speed of the right motor of the ship is increased by 40 percent, the speed of the left motor is reduced to 0, and the ship resumes direct voyage after the azimuth angle of the sailing obstacle becomes larger than 30 degrees on the starboard.
Further, the preset obstacle avoidance measures are carried out according to the following steps:
if the obstruction is a small float, this is done in the following way:
if the distance of the navigation obstacle is more than 10m, the azimuth angle of the navigation obstacle is within 15 degrees of the port; the left motor of the ship keeps increasing the speed by 20 percent, the speed of the right motor is reduced to 0, and the ship resumes direct voyage after the azimuth angle of the object in the way of voyage is increased to 30 degrees on the port;
if the distance of the navigation obstacle is more than 10 meters, the azimuth angle of the navigation obstacle is within 15 degrees of the starboard; the right motor of the ship keeps increasing the speed by 20 percent, the speed of the left motor is reduced to 0, and the ship resumes direct voyage after the azimuth angle of the object in the way of voyage is increased to 30 degrees on the starboard;
If the distance of the navigation obstacle is less than 10 meters, the azimuth angle of the navigation obstacle is within 15 degrees of the port; the left motor of the ship keeps increasing speed by 30%, the speed of the right motor is reduced to 0, and the ship resumes direct voyage after the azimuth angle of the object in the way is increased to 30 degrees on the port;
if the distance of the navigation obstacle is less than 10 meters, the azimuth angle of the navigation obstacle is within 15 degrees of the starboard; the right motor of the ship is kept to be increased by 30 percent, the speed of the left motor is reduced to 0, and the ship resumes straight voyage until the azimuth angle of the object in the way is increased to 30 degrees on the starboard.
Further, the preset obstacle avoidance measures are carried out according to the following steps:
if the obstruction is a medium-large float, this is done in the following way:
if the distance of the navigation obstacle is more than 30m, the azimuth angle of the navigation obstacle is within 15 degrees of the port; the left motor of the ship keeps increasing the speed by 20 percent, the speed of the right motor is reduced to 0, and the ship resumes direct voyage after the azimuth angle of the object in the way of voyage is increased to 30 degrees on the port;
if the distance of the navigation obstacle is more than 30 meters, the azimuth angle of the navigation obstacle is within 15 degrees of the starboard; the right motor of the ship keeps increasing the speed by 20 percent, the speed of the left motor is reduced to 0, and the ship resumes direct voyage after the azimuth angle of the object in the way of voyage is increased to 30 degrees on the starboard;
if the distance of the navigation obstacle is less than 30 meters, the azimuth angle of the navigation obstacle is within 15 degrees of the port; the left motor of the ship keeps increasing speed by 40%, the speed of the right motor is reduced to 0, and the ship resumes direct voyage after the azimuth angle of the object in the way is increased to 30 degrees on the port;
If the distance of the navigation obstacle is less than 30 meters, the azimuth angle of the navigation obstacle is within 15 degrees of the starboard; the right motor of the ship keeps increasing speed by 40 percent, the speed of the left motor is reduced to 0, and the ship resumes straight voyage after the azimuth angle of the object in the way is increased to 30 degrees on the starboard.
The invention provides an autonomous navigation control system of a channel unmanned ship based on target detection algorithm, which comprises a channel detection module, a router and an embedded system;
the embedded system is respectively connected with a channel detection module and a router which are arranged on the ship body;
the channel detection module is used for acquiring channel information and obtaining barrier information on the channel through the channel information;
the embedded system is used for processing and analyzing the channel image information to obtain the attribute information of the navigation obstacle and generating obstacle avoidance measures for controlling the unmanned ship according to the attribute information of the navigation obstacle, and the obstacle avoidance measures are as follows:
the navigation channel detection module collects whether navigation obstacles exist on the navigation channel through a scanning system arranged on the ship body, if the navigation obstacles exist, the distance between the ship body and the obstacles is calculated, if the distance is greater than a preset safety distance, the unmanned ship navigates autonomously, if the distance is less than or equal to the preset safety distance, an alarm signal is generated and sent to the far end, and the far end is reminded of manual control of the ship.
Further, the system comprises a ship body, a remote control terminal, a cloud server end, a power supply, a power system and a communication module, wherein the power supply, the power system and the communication module are arranged on the ship body;
the power system comprises a motor driving system; the motor is arranged on the ship body, and the motor drives the propeller pusher to enable the unmanned ship to move;
the communication module is arranged on the ship body and is in communication connection with the embedded system;
the cloud server is in communication connection with the communication module and is used for transmitting data;
the remote control terminal is in communication connection with the cloud server and is used for transmitting control instructions.
Further, the channel detection module comprises a laser radar and a visible light high-definition network camera;
the laser radar is used for detecting the navigation-blocking objects around the ship body, mainly the navigation-blocking objects in front of the ship body, the information of the detected navigation-blocking objects comprises the distance and the azimuth of the navigation-blocking objects relative to the ship, and the detected navigation-blocking object information is input into the ship control system;
the visible light high-definition network camera is used for imaging a real-time environment around a ship body, inputting formed image frames to target detection of the ship for automatically identifying categories and positions of navigation-blocking objects, and simultaneously carrying out plug flow service on the formed image frames by utilizing video streaming media service deployed at a cloud end so as to be capable of checking dynamic states around the ship in real time at an app end of a remote mobile phone.
The invention has the beneficial effects that:
according to the unmanned navigation path sweep control method and system based on the target detection algorithm, the unmanned navigation path sweep unmanned ship can realize specific type identification of the navigation-obstacle by utilizing the high-performance GPU embedded terminal carried by the unmanned navigation ship and the corresponding target detection algorithm, so that more accurate control is realized, and meanwhile, the laser radar is carried, and the distance measurement of the navigation-obstacle is more accurate. In addition, the ship route can be input at a remote control end at any time, the route is updated at any time, and the autonomous navigation of the unmanned ship is more accurate by combining the carried attitude instrument.
The unmanned ship of the present invention has two modes of motion: autonomous navigational mode and manual control mode. The remote monitoring technology can realize that the real-time environment around the ship can be checked on the remote mobile phone terminal app and the computer webpage terminal at the same time. In the manual control mode, a remote control instruction can be sent to the ship at a mobile phone end or a computer end, and manual control navigation is carried out on the ship; in the autonomous navigation mode, the ship can navigate autonomously according to the preset route sent by the remote mobile phone end or the computer end and combines corresponding obstacle avoidance measures, the preset route can select each route point at the mobile phone app end in a map point selection mode, and after each route point is selected, the route is automatically formed and sent to the ship end through the remote Cheng Tongxin.
Additional advantages, objects, and features of the invention will be set forth in part in the description which follows and in part will become apparent to those having ordinary skill in the art upon examination of the following or may be learned from practice of the invention. The objects and other advantages of the invention may be realized and obtained by means of the instrumentalities and combinations particularly pointed out in the specification.
Drawings
In order to make the objects, technical solutions and advantageous effects of the present invention more clear, the present invention provides the following drawings for description:
FIG. 1 is a schematic diagram of a unmanned ship for channel survey based on a target detection algorithm.
Fig. 2 is a schematic diagram of the composition and basic principle of the whole unmanned ship.
FIG. 3 is a flow chart of an autonomous navigational control algorithm.
FIG. 4 is a model of a mask-processed YOLOV3 target detection network.
Fig. 5 is a schematic diagram of a ship implementing corresponding motion control.
Fig. 6 is a manual control workflow diagram.
Fig. 7 is a schematic diagram of a state of a long-distance meeting or overtaking left turn in a preset obstacle avoidance measure.
Fig. 8 is a schematic diagram of a remote encounter or overtake right turn condition in a preset obstacle avoidance maneuver.
Fig. 9 is a schematic diagram of a close-up encounter or overtake left-turn condition in a preset obstacle avoidance measure.
Fig. 10 is a schematic diagram of a close-up encounter or overtake right turn condition in a preset obstacle avoidance maneuver.
In the figure, 1-laser radar, 2-front high-definition network camera, 3-right high-definition network camera, 4-left Gao Qingwang network camera, 5-4/5G router, 6-solar photovoltaic panel storage battery, 7-Jetson nano embedded board, 8-right motor and 9-left motor.
Detailed Description
The present invention will be further described with reference to the accompanying drawings and specific examples, which are not intended to limit the invention, so that those skilled in the art may better understand the invention and practice it.
As shown in fig. 1 and fig. 2, the unmanned ship for channel scanning based on the target detection algorithm provided in this embodiment includes a ship body, a power supply, a channel detection module, a router, an embedded system, and a power system, where the embedded system is connected with the channel detection module, the router, and the power system which are disposed on the ship body respectively; the power supply is respectively connected with the embedded system, the router and the power system; the channel detection module is used for acquiring channel information and obtaining barrier information on the channel through the channel information;
the channel detection module comprises a laser radar and a visible light high-definition network camera;
The laser radar is used for detecting the navigation-blocking objects around the ship body, mainly the navigation-blocking objects in front of the ship body, the information of the detected navigation-blocking objects comprises the distance and the azimuth of the navigation-blocking objects relative to the ship, and the detected navigation-blocking object information is input into the ship control system;
the visible light high-definition network camera is used for imaging a real-time environment around a ship body, inputting formed image frames to target detection of the ship for automatically identifying categories and positions of navigation-blocking objects, and simultaneously carrying out plug flow service on the formed image frames by utilizing video streaming media service deployed at a cloud end so as to be capable of checking dynamic states around the ship in real time at an app end of a remote mobile phone;
the embodiment also provides a Doppler depth finder, which is used for detecting the water depth of a water area navigated by a ship body and inputting the detected water depth point data into a ship communication system;
the embedded system adopts a high-performance GPU embedded terminal, is used for deploying a navigation obstacle target detection algorithm to automatically detect a navigation obstacle target in front of a ship body, deploying a communication system to receive a control instruction of a remote mobile phone app end, send alarm information and ship body position information to the remote mobile phone app end, and simultaneously upload acquired channel water depth point data to a cloud database, and is used for deploying a control system of the ship and performing navigation control on the ship according to input sensor information.
The power system comprises a motor driving system; the motor is arranged on the ship body, and the motor drives the propeller pusher to enable the unmanned ship to move;
the navigation channel detection module collects whether navigation obstacles exist on a navigation channel through a scanning system arranged on a ship body, if the navigation obstacles exist, the distance between the ship body and the obstacles is calculated, if the distance is greater than a preset safety distance, the unmanned ship navigates autonomously, and if the distance is less than or equal to the preset safety distance, an alarm signal is generated and sent to a far end, and the far end is reminded of manually controlling the ship;
the unmanned ship in the embodiment further comprises a communication module, a remote control terminal and a cloud server;
the communication module is arranged on the ship body and is in communication connection with the embedded system;
the cloud server is in communication connection with the communication module and is used for transmitting data;
the remote control terminal is in communication connection with the cloud server and is used for transmitting control instructions;
in this embodiment for the unmanned ship itself,
the laser radars are arranged in the central area of the front end of the ship body, and the number of the laser radars is 1;
the number of the visible light high-definition network cameras is 3, and the visible light high-definition network cameras are respectively arranged at the front end, the left side and the right side of the ship body and are used for acquiring the front view, the left view and the right view of the unmanned ship;
The number of the high-performance GPU embedded terminals is 1, and the high-performance GPU embedded terminals are used for data processing and control of the whole unmanned ship and are equivalent to the brain of the ship;
the number of the 4G/5G routers is 1, and the 4G/5G routers are used for constructing an unmanned ship local area network and simultaneously communicate with the outside;
the number of motors and motor drives is 2, and the motors and the motor drives are used for driving the unmanned ship to move;
the power supply is a storage battery, and a solar photovoltaic panel is used for storing energy;
the unmanned ship control module provided by the embodiment has two working modes, and is specifically as follows:
in the autonomous navigation mode, intelligent sensing is performed on the environment around the ship by mainly relying on a camera sensor, a laser radar sensor and a target detection algorithm deployed on the GPU embedded terminal, and then automatic control is performed on the ship according to sensing results;
and in the manual mode, manual control is carried out on the ship according to the received remote control instruction.
The high-performance GPU embedded terminal and the 3 high-definition network cameras are respectively connected with the 4G/5G router in a network mode, the laser radar and the Doppler depth finder are respectively connected with the high-performance GPU embedded terminal in a serial port mode, meanwhile, the high-performance GPU embedded terminal is connected with a motor drive, the motor drive is connected with a motor, and all the devices are powered by using a solar photovoltaic panel storage battery.
The remote control terminal is an android mobile phone terminal app;
the remote control terminal is in the form of android mobile phone app software and is used for checking the real-time environment around the unmanned ship and the real-time GPS position of the unmanned ship, meanwhile, a route can be preset for the unmanned ship, the unmanned ship can scan a channel according to the provided route, in addition, a remote control instruction can be sent to the unmanned ship, the ship can be remotely controlled when the unmanned ship encounters an emergency, and the safety of the unmanned ship is guaranteed.
For the cloud server, video streaming service, control signal transfer service, alarm information sending service and water depth point data storage are mainly provided for unmanned ships and remote control terminals.
The embedded system in the unmanned ship is further provided with an intelligent sensing module and an automatic control module; the intelligent perception module is used for automatically acquiring the environment around the ship body, and particularly comprises the step of automatically acquiring the type, the distance and the azimuth of the navigation obstacle in front of the ship body, wherein the adopted sensor comprises a high-definition camera and a laser radar, and the adopted algorithm comprises an improved yolov3 target detection algorithm;
the automatic control module is used for implementing the motion control of the ship according to the navigation obstacle information acquired by the sensing module, namely, accurately controlling the ship according to the acquired ship surrounding information, so that the ship can safely navigate;
As shown in fig. 3, fig. 3 is an autonomous navigation control algorithm flow, and the working process of the unmanned ship for channel sweep based on the target detection algorithm provided in this embodiment is as follows:
the navigation channel is scanned and the unmanned ship is placed at the river side, all the equipment is electrified and is in a waiting state, the real-time environment around the unmanned ship can be checked on a mobile phone app of a remote control terminal, and meanwhile, the real-time GPS position of the unmanned ship can be checked.
Under the autonomous navigation mode, before the unmanned ship sails, a route is set for the unmanned ship through the remote control terminal, the route is sent to the unmanned ship, and after the unmanned ship receives the preset route, the unmanned ship can start autonomous navigation and starts to execute the sweeping task of the route. The measured water depth data can be automatically uploaded to the cloud server database through the embedded terminal. When the unmanned ship encounters a navigation obstacle on a navigation line in the process of executing a navigation path sweep task in autonomous navigation, the unmanned ship can take obstacle avoidance measures by itself, and can return to a preset navigation line to continue navigation after confirming safety through a laser radar and a high-definition camera. The obstacle avoidance function adopted by the unmanned ship is automatically completed mainly by means of a carried laser radar, a high-definition camera and a deep learning target detection algorithm deployed in high-performance GPU embedded equipment, and the whole process is free from human intervention. The specific flow is as follows:
Collecting channel images through a high-definition camera;
performing image processing on the channel image to obtain a channel target class;
collecting channel laser images through a laser radar;
processing the channel laser image to obtain azimuth angles and distances of the channel navigation-obstacle and the navigation-obstacle;
judging whether the target exists or not, if not, carrying out direct voyage on the unmanned ship;
if yes, judging the type of the target or the navigation obstacle, and executing the preset obstacle avoidance measures according to the type of the navigation obstacle, wherein the preset obstacle avoidance measures provided by the embodiment are carried out according to the following steps:
the target types include any one or a combination of a navigation mark/buoy, a light boat, a cargo boat/mail ship, a small float, a medium and large float:
if the obstruction is a navigation mark/buoy, this is done in the following way:
if the distance of the navigation obstacle is more than 10m, the azimuth angle of the navigation obstacle is within 15 degrees of the port; the left motor of the ship keeps unchanged speed, the speed of the right motor is reduced to 0, and the ship resumes direct voyage after the azimuth angle of the object in the way of voyage is increased to 30 degrees on the port;
if the distance of the navigation obstacle is more than 10 meters, the azimuth angle of the navigation obstacle is within 15 degrees of the starboard; the right motor of the ship keeps unchanged speed, the speed of the left motor is reduced to 0, and the ship resumes direct voyage after the azimuth angle of the obstacle is increased to 30 degrees on the starboard;
If the distance of the navigation obstacle is less than 10 meters, the azimuth angle of the navigation obstacle is within 15 degrees of the port; executing the speed increase of the left motor of the ship by 20%, and reducing the speed of the right motor to 0 until the azimuth angle of the navigation-impaired object is increased to 30 degrees on the port side, and then restoring the direct navigation;
if the distance of the navigation obstacle is less than 10 meters, the azimuth angle of the navigation obstacle is within 15 degrees of the starboard; the speed of the right motor of the ship is increased by 20 percent, the speed of the left motor is reduced to 0, and the ship resumes straight voyage after the azimuth angle of the object in the way is increased to 30 degrees on the starboard.
In the preset obstacle avoidance measures provided in this embodiment, if the navigation obstacle is a lightship, the following manner is adopted:
if the distance of the navigation obstacle is more than 10m, the azimuth angle of the navigation obstacle is within 15 degrees of the port; the left motor of the ship keeps increasing the speed by 10 percent, the speed of the right motor is reduced to 0, and the ship resumes direct voyage after the azimuth angle of the object in the way of voyage is increased to 30 degrees on the port;
if the distance of the navigation obstacle is more than 10 meters, the azimuth angle of the navigation obstacle is within 15 degrees of the starboard; the right motor of the ship keeps increasing the speed by 10 percent, the speed of the left motor is reduced to 0, and the ship resumes direct voyage after the azimuth angle of the object in the way of voyage is increased to 30 degrees on the starboard;
if the distance of the navigation obstacle is less than 10 meters, the azimuth angle of the navigation obstacle is within 15 degrees of the port; the left motor of the ship keeps increasing speed by 30%, the speed of the right motor is reduced to 0, and the ship resumes direct voyage after the azimuth angle of the object in the way is increased to 30 degrees on the port;
If the distance of the navigation obstacle is less than 10 meters, the azimuth angle of the navigation obstacle is within 15 degrees of the starboard; the right motor of the ship is kept to be increased by 30 percent, the speed of the left motor is reduced to 0, and the ship resumes straight voyage until the azimuth angle of the object in the way is increased to 30 degrees on the starboard.
In the preset obstacle avoidance measures provided in this embodiment, if the obstacle is a cargo ship/mail ship, the following steps are performed:
if the distance of the obstacle is more than 30m, the azimuth angle of the obstacle is within 15 degrees of the port (namely, the azimuth angle of the obstacle is more than 0 degree of the port and less than 15 degrees of the port, and the situation is a front meeting state or a overtaking state); the left motor of the ship keeps increasing the speed by 20 percent, the speed of the right motor is reduced to 0, and the direct voyage is restored after the azimuth angle of the sailing obstacle becomes larger than 30 degrees of the port;
if the distance of the obstacle is more than 30 meters, the azimuth angle of the obstacle is within 15 degrees of starboard (namely, the azimuth angle of the obstacle is more than 0 degree of starboard and less than 15 degrees of starboard, and the situation is a front meeting state or a overtaking state); the right motor of the ship is kept at a speed increasing of 20%, the speed of the left motor is reduced to 0, and the direct voyage is restored after the azimuth angle of the sailing obstacle becomes larger than 30 degrees on the starboard;
if the space of the obstacle is less than 30 meters, the space angle of the obstacle is within 15 degrees of the port (namely, the space angle of the obstacle is greater than 0 degree of the port and less than 15 degrees of the port, and the situation is a front meeting state or a overtaking state); then the speed of the left motor of the ship is increased by 40%, the speed of the right motor is reduced to 0, and the ship resumes direct voyage after the azimuth angle of the sailing obstacle becomes larger than 30 degrees of the port;
If the distance of the obstacle is less than 30 meters, the azimuth angle of the obstacle is within 15 degrees of starboard (namely, the azimuth angle of the obstacle is greater than 0 degree of starboard and less than 15 degrees of starboard, and the situation is a front meeting state or a overtaking state); the speed of the right motor of the ship is increased by 40 percent, the speed of the left motor is reduced to 0, and the ship resumes direct voyage after the azimuth angle of the sailing obstacle becomes larger than 30 degrees on the starboard.
In the preset obstacle avoidance measures provided in this embodiment, if the obstacle is a small-sized floater, the following steps are performed:
if the distance of the navigation obstacle is more than 10m, the azimuth angle of the navigation obstacle is within 15 degrees of the port; the left motor of the ship keeps increasing the speed by 20 percent, the speed of the right motor is reduced to 0, and the ship resumes direct voyage after the azimuth angle of the object in the way of voyage is increased to 30 degrees on the port;
if the distance of the navigation obstacle is more than 10 meters, the azimuth angle of the navigation obstacle is within 15 degrees of the starboard; the right motor of the ship keeps increasing the speed by 20 percent, the speed of the left motor is reduced to 0, and the ship resumes direct voyage after the azimuth angle of the object in the way of voyage is increased to 30 degrees on the starboard;
if the distance of the navigation obstacle is less than 10 meters, the azimuth angle of the navigation obstacle is within 15 degrees of the port; the left motor of the ship keeps increasing speed by 30%, the speed of the right motor is reduced to 0, and the ship resumes direct voyage after the azimuth angle of the object in the way is increased to 30 degrees on the port;
If the distance of the navigation obstacle is less than 10 meters, the azimuth angle of the navigation obstacle is within 15 degrees of the starboard; the right motor of the ship is kept to be increased by 30 percent, the speed of the left motor is reduced to 0, and the ship resumes straight voyage until the azimuth angle of the object in the way is increased to 30 degrees on the starboard.
In the preset obstacle avoidance measures provided in this embodiment, if the obstacle is a medium-sized or large-sized floater, the following steps are performed:
if the distance of the navigation obstacle is more than 30m, the azimuth angle of the navigation obstacle is within 15 degrees of the port; the left motor of the ship keeps increasing the speed by 20 percent, the speed of the right motor is reduced to 0, and the ship resumes direct voyage after the azimuth angle of the object in the way of voyage is increased to 30 degrees on the port;
if the distance of the navigation obstacle is more than 30 meters, the azimuth angle of the navigation obstacle is within 15 degrees of the starboard; the right motor of the ship keeps increasing the speed by 20 percent, the speed of the left motor is reduced to 0, and the ship resumes direct voyage after the azimuth angle of the object in the way of voyage is increased to 30 degrees on the starboard;
if the distance of the navigation obstacle is less than 30 meters, the azimuth angle of the navigation obstacle is within 15 degrees of the port; the left motor of the ship keeps increasing speed by 40%, the speed of the right motor is reduced to 0, and the ship resumes direct voyage after the azimuth angle of the object in the way is increased to 30 degrees on the port;
if the distance of the navigation obstacle is less than 30 meters, the azimuth angle of the navigation obstacle is within 15 degrees of the starboard; the right motor of the ship keeps increasing speed by 40 percent, the speed of the left motor is reduced to 0, and the ship resumes straight voyage after the azimuth angle of the object in the way is increased to 30 degrees on the starboard.
The ship obstacle avoidance measures meeting the international ship collision avoidance rule provided in this embodiment may be implemented in the following manner:
if the distance of the obstacle is more than 30m, the azimuth angle of the obstacle is within 15 degrees of the port (namely, the azimuth angle of the obstacle is more than 0 degree of the port and less than 15 degrees of the port, and the situation is a front meeting state or a overtaking state); the left motor of the ship keeps increasing the speed by 20 percent, the speed of the right motor is reduced to 0, and the direct voyage is restored after the azimuth angle of the sailing obstacle becomes larger than 30 degrees of the port; as shown in fig. 7.
If the distance of the obstacle is more than 30 meters, the azimuth angle of the obstacle is within 15 degrees of starboard (namely, the azimuth angle of the obstacle is more than 0 degree of starboard and less than 15 degrees of starboard, and the situation is a front meeting state or a overtaking state); the right motor of the ship is kept at a speed increasing of 20%, the speed of the left motor is reduced to 0, and the direct voyage is restored after the azimuth angle of the sailing obstacle becomes larger than 30 degrees on the starboard; as shown in fig. 8;
if the space of the obstacle is less than 30 meters, the space angle of the obstacle is within 15 degrees of the port (namely, the space angle of the obstacle is greater than 0 degree of the port and less than 15 degrees of the port, and the situation is a front meeting state or a overtaking state); then the speed of the left motor of the ship is increased by 40%, the speed of the right motor is reduced to 0, and the ship resumes direct voyage after the azimuth angle of the sailing obstacle becomes larger than 30 degrees of the port; as shown in fig. 9.
If the distance of the obstacle is less than 30 meters, the azimuth angle of the obstacle is within 15 degrees of starboard (namely, the azimuth angle of the obstacle is greater than 0 degree of starboard and less than 15 degrees of starboard, and the situation is a front meeting state or a overtaking state); then the speed of the right motor of the ship is increased by 40%, the speed of the left motor is reduced to 0, and the ship resumes direct voyage after the azimuth angle of the sailing obstacle becomes larger than 30 degrees on the starboard; as shown in fig. 10.
As shown in fig. 4, fig. 4 is a YOLOV3 target detection network model of mask processing, specifically, a deep learning target detection algorithm deployed in a high-performance GPU embedded device, which is a YOLOV3 target detection network model fused with a transducer structure, and the model can automatically, quickly and accurately detect types of navigation obstacles of a common type in front of a ship, and the embedded system in the embodiment is further provided with an improved YOLOV3 target detection network model, where the improved YOLOV3 target detection network model is implemented according to the following steps:
the improved YOLOV3 target detection network model provided in this embodiment is as follows:
acquiring channel image data shot by a channel;
preprocessing an image to obtain a plurality of image blocks;
masking all the image blocks by using a masking method according to a uniform distribution function, wherein the number of the masked image blocks in the masking process is determined according to a preset value and is generally selected to be 20% -50%;
And inputting the image data processed by the mask method into a backbone network for feature extraction, and acquiring features with different scales by adopting a method of an original yolov3 network model in the feature extraction process to finally obtain the navigation obstacle target.
The specific process of the YOLOV3 target detection network model provided in this embodiment is: inputting a ship front view map shot by a high-definition camera, changing the size of the ship front view map into 416 x 416 pixel points after image preprocessing, dividing the preprocessed image into 16 x 16 equal-sized image blocks, masking all the image blocks by using a masking method according to a uniform distribution function, wherein the number of the masked image blocks accounts for 50% of the number of all the image blocks, acquiring a 3-channel feature map corresponding to the image processed by using the masking method, inputting the feature map into a backbone network (backbone) for feature extraction, and acquiring 3 features with different scales by using a method of an original yolov3 network model in the feature extraction process, thereby improving the detection capability of a small target. The improved yolov3 target detection model provided in the embodiment adopts a masking method on the basis of the original yolov3 model, and the improved yolov3 target detection model can reduce over-fitting, enhance the robustness of the model and improve the accuracy of target detection. The specific masking method is described as follows:
The input image is divided into areas, the original image is divided into a plurality of equal-sized small image blocks, then the image blocks are subjected to shielding processing according to a uniform distribution function, and when the number of the shielded image blocks accounts for 20%, 30% or 50% of the number of all the image blocks, for example, the shielding proportion is 50%, half of the area of the original image is invisible. Because most of the background area of the image collected by the ship sailing on the water surface is water and sky, the object of the navigation obstacle only occupies a small part of the whole image, the characteristics of the water are basically similar, and the characteristics of the sky are basically similar, that is, most of the image blocks where the water is located and the image blocks where the sky is located are redundant, therefore, the redundancy of the extracted characteristics of the network can be greatly reduced, the fitting is reduced, and finally the detection precision of the object of the navigation obstacle is improved.
When the unmanned ship encounters an obstacle, the unmanned ship can acquire the surrounding environment through the high-definition camera, then the corresponding video frame is input into the GPU embedded equipment carried by the ship, the type of the obstacle in the image frame is automatically identified by using the deployed deep learning target detection algorithm, the position of the obstacle in the image is recorded, meanwhile, the carried laser radar can automatically detect the obstacle, the distance and the azimuth information of the obstacle are recorded, the unmanned ship is automatically combined with the image identification result for analysis, the laser radar is only used for measuring the distance and the azimuth of the obstacle relative to the ship, the result of identifying the image of the camera through the target detection algorithm comprises the specific type of the obstacle in the image and the specific coordinate position of the obstacle in the image, and the accurate identification of the obstacle information can be realized by comparing the specific type of the obstacle with the azimuth information of the laser radar, so that the attribute of the obstacle (the specific attribute comprises the type of the obstacle, the distance and the azimuth of the obstacle relative to the ship) is accurately judged, and the ship is accurately controlled to implement the corresponding obstacle avoidance.
The pseudo code of the control algorithm for precisely controlling the ship obstacle avoidance is as follows:
/>
the control algorithm has the advantages that: the algorithm has strong pertinence, and is simple and accurate.
Compared with the original YOLOV3 target detection algorithm, the YOLOV3 target detection algorithm of the masking process in this embodiment performs masking processing on the original input image, as described above, when the model is trained, the input image is divided into blocks, the original input image is divided into a plurality of equal-sized blocks, then the image blocks are randomly masked in a certain proportion, and then the masked image is input into a backstone for further processing. Therefore, the robustness of the model is greatly improved, and the accuracy of the model is improved. Especially for the target detection on the water surface, as the unmanned ship sailing on the water surface has the background parts of the visual field image mostly of water and sky, and the background parts have serious feature redundancy in the forward propagation process of the model, the robustness of the model can be greatly improved after the improved yolov3 algorithm of the embodiment.
When the unmanned ship still cannot guarantee the sailing safety by means of the obstacle avoidance measures of the unmanned ship, alarm information can be automatically sent to the remote control terminal, workers of the remote control terminal are reminded of taking over the unmanned ship to control the unmanned ship, after the workers receive the alarm information, the workers can check the real-time environment around the unmanned ship and the real-time GPS position through the remote control terminal, and control instructions are sent to the unmanned ship through the remote control function, so that the ship is operated, and the ship is in a safe environment.
In the process, the unmanned ship receives a preset route sent by the mobile phone app end and sends alarm information to the mobile phone app end, and the adopted remote communication technology is a remote communication method designed based on the MQTT protocol and the cloud server. In the manual mode, a control command is sent from the mobile phone app end to the unmanned ship through the remote communication method. The method is high in real-time performance, stable and reliable, and the remote communication method can be used as long as 4G/5G signals exist in the water area where the unmanned ship sails. When the unmanned ship autonomous navigation execution task sails to the place without the 4G/5G signal, the unmanned ship can sail according to the preset route and autonomous navigation capacity on the unmanned ship control panel.
As shown in fig. 5, in the manual mode, the remote mobile phone app end can send corresponding speed control instructions to two motors on the unmanned ship, and the ship end performs corresponding motion control on the ship according to the received remote control instructions, and specifically, the remote control module includes five operations in total: the left motor stepless speed regulation (left progress bar of the lower end control part on the figure), the right motor stepless speed regulation (right progress bar of the lower end control part on the figure), the two motors stepless speed regulation at the same time (middle progress bar of the lower end control part on the figure), starting, reversing and stopping. Through the five operation controls, the left turn, the right turn, the straight running, the reversing and the stopping of the ship can be flexibly realized at various speeds.
As shown in fig. 6, a specific manual control workflow is as follows:
when the unmanned ship still cannot guarantee the sailing safety by means of the obstacle avoidance measures of the unmanned ship, alarm information can be automatically sent to the remote control terminal, workers of the remote control terminal are reminded of taking over the unmanned ship to control the unmanned ship, after the workers receive the alarm information, the workers can check the real-time environment around the unmanned ship and the real-time GPS position through the remote control terminal, and control instructions are sent to the unmanned ship through the remote control function, so that the ship is operated, and the ship is in a safe environment.
The unmanned ship of the present embodiment has two movement patterns: autonomous navigational mode and manual control mode. The remote monitoring technology of the embodiment can realize that the real-time environment around the ship can be checked on the remote mobile phone terminal app and the computer webpage terminal at the same time. In the manual control mode, a remote control instruction can be sent to the ship at a mobile phone end or a computer end, and manual control navigation is carried out on the ship; in the autonomous navigation mode, the ship can perform autonomous navigation according to a preset route sent by a remote mobile phone end or a computer end and combined with corresponding obstacle avoidance measures, wherein the preset route can select each route point at the mobile phone app end in a map point selection mode, and after each route point is selected, the route is automatically formed and sent to the ship end through a remote Cheng Tongxin.
The above-described embodiments are merely preferred embodiments for fully explaining the present invention, and the scope of the present invention is not limited thereto. Equivalent substitutions and modifications will occur to those skilled in the art based on the present invention, and are intended to be within the scope of the present invention. The protection scope of the invention is subject to the claims.
Claims (10)
1. The autonomous navigation control method of the unmanned ship for channel sweep based on the target detection algorithm is characterized by comprising the following steps: the method comprises the following steps:
acquiring channel surrounding environment data through a camera, inputting the channel surrounding environment data into a GPU embedded device carried by a ship, identifying the type of the navigation obstacle in an image frame by using a deployed deep learning target detection algorithm, and recording the position of the navigation obstacle in the image;
the navigation obstacle is detected through the laser radar, the distance and azimuth information of the navigation obstacle are recorded, the combination analysis is carried out on the distance and azimuth information and the image recognition result, the attribute information of the navigation obstacle is obtained, and unmanned ship navigation preset obstacle avoidance measures are adopted according to the obtained attribute information of the navigation obstacle.
2. The autonomous navigation control method of the unmanned ship for channel surveying based on the target detection algorithm as claimed in claim 1, wherein: the attribute information of the navigation obstacle comprises the category of the navigation obstacle, the distance and the azimuth of the navigation obstacle and the unmanned ship.
3. The autonomous navigation control method of the unmanned ship for channel surveying based on the target detection algorithm as claimed in claim 1, wherein: the preset obstacle avoidance measures are carried out according to the following steps:
if the obstruction is a navigation mark/buoy, this is done in the following way:
if the distance of the navigation obstacle is more than 10m, the azimuth angle of the navigation obstacle is within 15 degrees of the port; the left motor of the ship keeps unchanged speed, the speed of the right motor is reduced to 0, and the ship resumes direct voyage after the azimuth angle of the object in the way of voyage is increased to 30 degrees on the port;
if the distance of the navigation obstacle is more than 10 meters, the azimuth angle of the navigation obstacle is within 15 degrees of the starboard; the right motor of the ship keeps unchanged speed, the speed of the left motor is reduced to 0, and the ship resumes direct voyage after the azimuth angle of the obstacle is increased to 30 degrees on the starboard;
if the distance of the navigation obstacle is less than 10 meters, the azimuth angle of the navigation obstacle is within 15 degrees of the port; executing the speed increase of the left motor of the ship by 20%, and reducing the speed of the right motor to 0 until the azimuth angle of the navigation-impaired object is increased to 30 degrees on the port side, and then restoring the direct navigation;
if the distance of the navigation obstacle is less than 10 meters, the azimuth angle of the navigation obstacle is within 15 degrees of the starboard; the speed of the right motor of the ship is increased by 20 percent, the speed of the left motor is reduced to 0, and the ship resumes straight voyage after the azimuth angle of the object in the way is increased to 30 degrees on the starboard.
4. The autonomous navigation control method of the unmanned ship for channel surveying based on the target detection algorithm as claimed in claim 1, wherein: the preset obstacle avoidance measures are carried out according to the following steps:
if the obstruction is a lightboat, this is done in the following way:
if the distance of the navigation obstacle is more than 10m, the azimuth angle of the navigation obstacle is within 15 degrees of the port; the left motor of the ship keeps increasing the speed by 10 percent, the speed of the right motor is reduced to 0, and the ship resumes direct voyage after the azimuth angle of the object in the way of voyage is increased to 30 degrees on the port;
if the distance of the navigation obstacle is more than 10 meters, the azimuth angle of the navigation obstacle is within 15 degrees of the starboard; the right motor of the ship keeps increasing the speed by 10 percent, the speed of the left motor is reduced to 0, and the ship resumes direct voyage after the azimuth angle of the object in the way of voyage is increased to 30 degrees on the starboard;
if the distance of the navigation obstacle is less than 10 meters, the azimuth angle of the navigation obstacle is within 15 degrees of the port; the left motor of the ship keeps increasing speed by 30%, the speed of the right motor is reduced to 0, and the ship resumes direct voyage after the azimuth angle of the object in the way is increased to 30 degrees on the port;
if the distance of the navigation obstacle is less than 10 meters, the azimuth angle of the navigation obstacle is within 15 degrees of the starboard; the right motor of the ship is kept to be increased by 30 percent, the speed of the left motor is reduced to 0, and the ship resumes straight voyage until the azimuth angle of the object in the way is increased to 30 degrees on the starboard.
5. The autonomous navigation control method of the unmanned ship for channel surveying based on the target detection algorithm as claimed in claim 1, wherein: the preset obstacle avoidance measures are carried out according to the following steps:
if the obstacle is a cargo ship/mail ship, this is done in the following way:
if the distance of the obstacle is more than 30m, the azimuth angle of the obstacle is within 15 degrees of the port (namely, the azimuth angle of the obstacle is more than 0 degree of the port and less than 15 degrees of the port, and the situation is a front meeting state or a overtaking state); the left motor of the ship keeps increasing the speed by 20 percent, the speed of the right motor is reduced to 0, and the direct voyage is restored after the azimuth angle of the sailing obstacle becomes larger than 30 degrees of the port;
if the distance of the obstacle is more than 30 meters, the azimuth angle of the obstacle is within 15 degrees of starboard (namely, the azimuth angle of the obstacle is more than 0 degree of starboard and less than 15 degrees of starboard, and the situation is a front meeting state or a overtaking state); the right motor of the ship is kept at a speed increasing of 20%, the speed of the left motor is reduced to 0, and the direct voyage is restored after the azimuth angle of the sailing obstacle becomes larger than 30 degrees on the starboard;
if the space of the obstacle is less than 30 meters, the space angle of the obstacle is within 15 degrees of the port (namely, the space angle of the obstacle is greater than 0 degree of the port and less than 15 degrees of the port, and the situation is a front meeting state or a overtaking state); then the speed of the left motor of the ship is increased by 40%, the speed of the right motor is reduced to 0, and the ship resumes direct voyage after the azimuth angle of the sailing obstacle becomes larger than 30 degrees of the port;
If the distance of the obstacle is less than 30 meters, the azimuth angle of the obstacle is within 15 degrees of starboard (namely, the azimuth angle of the obstacle is greater than 0 degree of starboard and less than 15 degrees of starboard, and the situation is a front meeting state or a overtaking state); the speed of the right motor of the ship is increased by 40 percent, the speed of the left motor is reduced to 0, and the ship resumes direct voyage after the azimuth angle of the sailing obstacle becomes larger than 30 degrees on the starboard.
6. The autonomous navigation control method of the unmanned ship for channel surveying based on the target detection algorithm as claimed in claim 1, wherein: the preset obstacle avoidance measures are carried out according to the following steps:
if the obstruction is a small float, this is done in the following way:
if the distance of the navigation obstacle is more than 10m, the azimuth angle of the navigation obstacle is within 15 degrees of the port; the left motor of the ship keeps increasing the speed by 20 percent, the speed of the right motor is reduced to 0, and the ship resumes direct voyage after the azimuth angle of the object in the way of voyage is increased to 30 degrees on the port;
if the distance of the navigation obstacle is more than 10 meters, the azimuth angle of the navigation obstacle is within 15 degrees of the starboard; the right motor of the ship keeps increasing the speed by 20 percent, the speed of the left motor is reduced to 0, and the ship resumes direct voyage after the azimuth angle of the object in the way of voyage is increased to 30 degrees on the starboard;
if the distance of the navigation obstacle is less than 10 meters, the azimuth angle of the navigation obstacle is within 15 degrees of the port; the left motor of the ship keeps increasing speed by 30%, the speed of the right motor is reduced to 0, and the ship resumes direct voyage after the azimuth angle of the object in the way is increased to 30 degrees on the port;
If the distance of the navigation obstacle is less than 10 meters, the azimuth angle of the navigation obstacle is within 15 degrees of the starboard; the right motor of the ship is kept to be increased by 30 percent, the speed of the left motor is reduced to 0, and the ship resumes straight voyage until the azimuth angle of the object in the way is increased to 30 degrees on the starboard.
7. The autonomous navigation control method of the unmanned ship for channel surveying based on the target detection algorithm as claimed in claim 1, wherein: the preset obstacle avoidance measures are carried out according to the following steps:
if the obstruction is a medium-large float, this is done in the following way:
if the distance of the navigation obstacle is more than 30m, the azimuth angle of the navigation obstacle is within 15 degrees of the port; the left motor of the ship keeps increasing the speed by 20 percent, the speed of the right motor is reduced to 0, and the ship resumes direct voyage after the azimuth angle of the object in the way of voyage is increased to 30 degrees on the port;
if the distance of the navigation obstacle is more than 30 meters, the azimuth angle of the navigation obstacle is within 15 degrees of the starboard; the right motor of the ship keeps increasing the speed by 20 percent, the speed of the left motor is reduced to 0, and the ship resumes direct voyage after the azimuth angle of the object in the way of voyage is increased to 30 degrees on the starboard;
if the distance of the navigation obstacle is less than 30 meters, the azimuth angle of the navigation obstacle is within 15 degrees of the port; the left motor of the ship keeps increasing speed by 40%, the speed of the right motor is reduced to 0, and the ship resumes direct voyage after the azimuth angle of the object in the way is increased to 30 degrees on the port;
If the distance of the navigation obstacle is less than 30 meters, the azimuth angle of the navigation obstacle is within 15 degrees of the starboard; the right motor of the ship keeps increasing speed by 40 percent, the speed of the left motor is reduced to 0, and the ship resumes straight voyage after the azimuth angle of the object in the way is increased to 30 degrees on the starboard.
8. The autonomous navigation control system of the unmanned ship for channel sweep based on the target detection algorithm is characterized in that: the system comprises a channel detection module, a router and an embedded system;
the embedded system is respectively connected with a channel detection module and a router which are arranged on the ship body;
the channel detection module is used for acquiring channel information and obtaining barrier information on the channel through the channel information;
the embedded system is used for processing and analyzing the channel image information to obtain the attribute information of the navigation obstacle and generating obstacle avoidance measures for controlling the unmanned ship according to the attribute information of the navigation obstacle, and the obstacle avoidance measures are as follows:
the navigation channel detection module collects whether navigation obstacles exist on the navigation channel through a scanning system arranged on the ship body, if the navigation obstacles exist, the distance between the ship body and the obstacles is calculated, if the distance is greater than a preset safety distance, the unmanned ship navigates autonomously, if the distance is less than or equal to the preset safety distance, an alarm signal is generated and sent to the far end, and the far end is reminded of manual control of the ship.
9. The unmanned navigation path sweep vessel autonomous navigational control system based on the target detection algorithm according to claim 8, wherein: the system also comprises a ship body, a remote control terminal, a cloud server end, a power supply, a power system and a communication module which are arranged on the ship body;
the power system comprises a motor driving system; the motor is arranged on the ship body, and the motor drives the propeller pusher to enable the unmanned ship to move;
the communication module is arranged on the ship body and is in communication connection with the embedded system;
the cloud server is in communication connection with the communication module and is used for transmitting data;
the remote control terminal is in communication connection with the cloud server and is used for transmitting control instructions.
10. The unmanned navigation path sweep vessel autonomous navigational control system based on the target detection algorithm according to claim 8, wherein: the channel detection module comprises a laser radar and a visible light high-definition network camera;
the laser radar is used for detecting the navigation-blocking objects around the ship body, mainly the navigation-blocking objects in front of the ship body, the information of the detected navigation-blocking objects comprises the distance and the azimuth of the navigation-blocking objects relative to the ship, and the detected navigation-blocking object information is input into the ship control system;
The visible light high-definition network camera is used for imaging a real-time environment around a ship body, inputting formed image frames to target detection of the ship for automatically identifying categories and positions of navigation-blocking objects, and simultaneously carrying out plug flow service on the formed image frames by utilizing video streaming media service deployed at a cloud end so as to be capable of checking dynamic states around the ship in real time at an app end of a remote mobile phone.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310983281.6A CN116859948A (en) | 2023-08-04 | 2023-08-04 | Autonomous navigation control method and system for unmanned ship for channel sweep based on target detection algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310983281.6A CN116859948A (en) | 2023-08-04 | 2023-08-04 | Autonomous navigation control method and system for unmanned ship for channel sweep based on target detection algorithm |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116859948A true CN116859948A (en) | 2023-10-10 |
Family
ID=88225084
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310983281.6A Pending CN116859948A (en) | 2023-08-04 | 2023-08-04 | Autonomous navigation control method and system for unmanned ship for channel sweep based on target detection algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116859948A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117698943A (en) * | 2023-12-27 | 2024-03-15 | 重庆交通大学 | Intelligent navigation mark ship and risk detection method |
-
2023
- 2023-08-04 CN CN202310983281.6A patent/CN116859948A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117698943A (en) * | 2023-12-27 | 2024-03-15 | 重庆交通大学 | Intelligent navigation mark ship and risk detection method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Zhang et al. | Subsea pipeline leak inspection by autonomous underwater vehicle | |
| Heidarsson et al. | Obstacle detection and avoidance for an autonomous surface vehicle using a profiling sonar | |
| CN110580044A (en) | unmanned ship full-automatic navigation heterogeneous system based on intelligent sensing | |
| CN102306025A (en) | Wireless remote-control automatic sampling monitoring ship | |
| KR102339465B1 (en) | Autonomous navigation ship system for removing sea waste based on deep learning-vision recognition | |
| CN109515086A (en) | Hydrospace detection robot and its operational method | |
| CN209426501U (en) | Hydrospace detection robot | |
| CN105184816A (en) | Visual inspection and water surface target tracking system based on USV and detection tracking method thereof | |
| CN109835441B (en) | Automatic driving method and system for water quality monitoring intelligent boat | |
| CN108227739B (en) | Close-range obstacle avoidance method of underwater automatic driving equipment and underwater automatic driving equipment | |
| CN205539467U (en) | Unmanned ship's navigation environment barrier recognition device | |
| CN116859948A (en) | Autonomous navigation control method and system for unmanned ship for channel sweep based on target detection algorithm | |
| CN113110514A (en) | Unmanned ship navigation obstacle avoidance system and method based on big data | |
| KR20200090565A (en) | Smart underwater drone system with third party point-of-view camera | |
| CN109240315A (en) | A kind of underwater automatic obstacle avoiding system and underwater barrier-avoiding method | |
| KR20220055555A (en) | Method and device for monitoring harbor and ship | |
| CN115019412A (en) | Underwater AUV (autonomous underwater vehicle) submarine cable inspection system and method based on multiple sensors | |
| CN202120123U (en) | Automatic wireless remote-control sampling monitoring boat | |
| CN109541609A (en) | Rivers and lakes sniffing robot and its operational method | |
| CN113625726A (en) | Unmanned ship obstacle avoidance method and system | |
| CN212460059U (en) | Water surface radioactive unmanned ship monitoring device | |
| CN112829885A (en) | Unmanned autonomous reconnaissance ship | |
| CN209728169U (en) | Rivers and lakes sniffing robot | |
| Jorge et al. | VITA1: An unmanned underwater vehicle prototype for operation in underwater tunnels | |
| CN111897342A (en) | Automatic berthing and departing system and method for modern ship |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |