A multi-amphibious unmanned platform and its motion control method

Abstract

This invention provides an amphibious unmanned platform and its motion control method, belonging to the field of amphibious unmanned platforms. It addresses the problems of poor obstacle-crossing performance, slippage, poor scalability, poor environmental adaptability, and poor maneuverability found in existing amphibious vehicles. The platform includes an upper platform, an underwater propulsion device, and a land-based power unit. The bow and stern horizontal thrusters work together to enable the amphibious unmanned transport platform to turn in place, significantly reducing the turning radius. The bow and stern vertical thrusters work together to adjust the pitch angle of the amphibious unmanned transport platform, facilitating rapid submersion. Tracks enhance the platform's underwater and land obstacle-crossing capabilities, adapting to various terrain environments. The amphibious unmanned platform also employs wireless control, autonomously maneuvering according to a predetermined program, reducing its limitations during movement. Furthermore, it boasts advantages such as light weight, long remote control distance, long operating time, high efficiency, compact structure, and convenient operation.

Description

Technical Field

[0001] This invention relates to the field of amphibious unmanned platform technology, and more specifically, to an amphibious unmanned platform and its motion control method. Background Technology

[0002] With the development of the world economy and the needs of national defense, vigorously developing the marine economy, exploiting marine resources, and strengthening marine military capabilities have become an inevitable trend. However, with the deepening of marine development and the further demands of military buildup, there is a need for a multi-functional unmanned transport platform that can operate in environments inaccessible to humans or in hazardous environments, equipped with various functional devices, capable of both underwater navigation and underwater crawling, and terrestrial movement, to perform tasks such as marine environmental observation, seabed mapping, underwater rescue, unmanned island landing operations, and military reconnaissance.

[0003] Currently, ordinary robots can only move in water or on land and cannot switch between amphibious movements, resulting in poor comprehensive operational capabilities. Based on the current development status at home and abroad, amphibious vehicles, mainly biomimetic and propeller-driven, suffer from poor obstacle crossing performance, easy slippage, poor scalability, poor environmental adaptability, and poor maneuverability. Summary of the Invention

[0004] The technical problem to be solved by this invention is:

[0005] In order to solve the problems of poor obstacle crossing performance, easy slippage, poor scalability, poor environmental adaptability and poor maneuverability of existing multi-purpose vehicles.

[0006] The technical solution adopted by the present invention to solve the above-mentioned technical problems is as follows:

[0007] This invention provides an amphibious unmanned platform, comprising an upper platform, an underwater propulsion device, and a land-based power unit.

[0008] The upper platform includes an enclosed electronic cabin and an external frame outside the electronic cabin. The electronic cabin is located at the center of the external frame and contains a controlled system, a navigation and positioning system, a communication system, and a power control system. The external frame is equipped with an underwater propulsion system, and the upper platform is equipped with a land-based power unit below it.

[0009] The underwater propulsion system includes a left main thruster and a right main thruster symmetrically arranged on both sides of the central axis of the stern of the multi-purpose unmanned platform. It also includes a stern vertical thruster, a stern horizontal thruster, a bow vertical thruster, and a bow horizontal thruster. The propellers of the left and right main thrusters are horizontally oriented towards the rear of the multi-purpose unmanned platform. The stern vertical thruster is located at the bottom of the external frame, and its propeller faces the tracked chassis of the land-based power unit. The stern horizontal thruster and the bow horizontal thruster are parallel to each other and located on opposite sides of the bow and stern of the multi-purpose unmanned platform. The propeller of the bow vertical thruster faces a direction perpendicular to the port side, and the propeller of the bow horizontal thruster faces a direction perpendicular to the starboard side.

[0010] The land-based power system includes a tracked chassis and two track assemblies located below the tracked chassis. Each track assembly includes a drive wheel, a guide wheel, a load wheel, a track, and a track motor. Several load wheels are evenly distributed between the drive wheel and the guide wheel, and the drive wheel and the guide wheel are located diagonally above the load wheels at both ends. The track wraps around the drive wheel, the guide wheel, and the several load wheels. The output end of the track motor is connected to the drive wheel. The drive wheel, the guide wheel, and the load wheels are all rotatably connected to the tracked chassis.

[0011] Furthermore, the controlled system includes a motion computer, a mission computer, and a guidance computer. The motion computer is used for multi-sensor data processing and for calculating pose control commands for the multi-amphibious unmanned platform. The mission computer is used for mission planning, network communication, and fault monitoring. The guidance computer is used to provide guidance commands, including desired heading, desired speed, and desired depth.

[0012] Furthermore, the navigation and positioning system includes a GPS module and an inertial navigation sensor inside the electronic cabin, as well as a Doppler sensor and a depth gauge located outside the electronic cabin; used for positioning and attitude calculation of the amphibious unmanned platform; the communication system includes a WiFi communication module and a wireless communication module, used for the amphibious unmanned platform to connect to the host computer through the communication system.

[0013] Furthermore, the power control system includes a main microcontroller control board and a relay matrix board. Each pin of the main microcontroller control board is used to control a relay switch, thereby controlling the power supply of each module. The electronic cabin adopts dual power supply, which is a battery.

[0014] Furthermore, several buoyancy blocks are provided above the tracked chassis.

[0015] Furthermore, the track motor is powered by a battery installed on the track chassis, and the track chassis is equipped with a watertight control compartment. The control compartment contains a track microcontroller, which is used to control the track motor and communicate with the controlled system via a wired network.

[0016] Furthermore, for the controlled system, a PID controller tuned based on an RBF neural network is set up, including:

[0017] Let X = [x1, x2, ..., x n ] T Given the network input vector, the radial basis vectors of the hidden layer nodes are H = [h1, h2, ..., h...]. m ] T h j For Gaussian functions:

[0018]

[0019] The center vector of the j-th node in the hidden layer of the RBF neural network is C. j =[c j1 ,c j2 ,...,c jm ] T Let the basis width vector of the hidden layer nodes in the RBF neural network be B = [b1, b2, ..., b m ] T , where b j Let w be the base width parameter of hidden layer node j, and it is a number greater than 0; the weight vector of the network is w = [w1, w2, ..., w...]. n ] T The network output for identification is:

[0020] y mout =w1h1+w2h2+...+w m h m

[0021] Let the theoretical output of the system to be identified at time k be y. out (k), the output of the network identification is y mout (k), then the system performance index function is:

[0022]

[0023] The output weights, node centers, and node base width parameters are obtained using the gradient descent method.

[0024] The output weight expression is as follows:

[0025] w j (k)=w j (k-1)+η(y out (k)-y mout (k))h j +α(w j (k-1)-w j (k-2))+β(wj (k-2)-w j (k-3) where η is the learning rate; α and β are momentum factors, both ranging from (0, 1);

[0026] Node base width variation Δb j for:

[0027]

[0028] Node base width parameter b j for:

[0029] b j (k)=b j (k-1)+ηΔb j +α(b j (k-1)-b j (k-2))+β(b j (k-2)-b j (k-3))

[0030] Node center change Δc ji for:

[0031]

[0032] Node center c j for:

[0033] c ji =c ji (k-1)+ηΔc ji +α(c ji (k-1)-c ji (k-2))+β(c ji (k-2)-c ji (k-3))

[0034] Where cj = [c11, c12, c13…c1m];

[0035] The PID parameter tuning algorithm is as follows:

[0036] The control error is:

[0037] error(k) = u(k) - yout(k)

[0038] Where u(k) is the input of the controlled system and yout(k) is the output of the controlled system;

[0039] The three PID inputs are:

[0040] xc(1) = error(k) - error(k-1)

[0041] xc(2) = error(k)

[0042] xc(3)=error(k)-2error(k-1)+error(k-2)

[0043] The neural network tuning index is:

[0044]

[0045] The three parameters K are obtained by gradient descent. p ,K I ,K D The adjustment amount in decibels is:

[0046]

[0047]

[0048]

[0049] in, The sensitivity of the object's output to changes in the control input; K p For the proportional gain term, K i K is the integral gain term. d This is the differential gain term;

[0050]

[0051] This invention discloses a motion control method for an amphibious unmanned platform, comprising the following steps:

[0052] S100. When the amphibious unmanned platform moves in water, it is driven only by the propulsion system. The motion is decomposed into horizontal and vertical motion using underactuated control. The thrust of the bow and stern vertical thrusters is controlled by the controlled system to keep the amphibious unmanned platform at a constant depth. The bow and navigation of the amphibious unmanned platform are controlled by the thrust of the bow and stern horizontal thrusters. The forward speed of the amphibious unmanned platform is changed by controlling the thrust output of the thrusters.

[0053] S200: When the multi-purpose unmanned platform moves on land, it is driven only by the land-based power unit. The differential speed of the tracks is controlled by the microcontroller in the control cabin set on the tracked chassis to realize the turning of the multi-purpose unmanned platform on land. The speed of the tracks is controlled by controlling the duty cycle of the track motor to realize the acceleration, deceleration, forward and backward maneuvering modes.

[0054] S300: When the multi-purpose unmanned platform moves underwater, it uses a combination of propulsion system and land-based power unit for drive. By controlling the bow and stern vertical thrusters to apply downward thrust, the pressure between the multi-purpose unmanned platform and the seabed is increased. The left and right main thrusters are controlled to assist the multi-purpose unmanned platform in moving underwater. The differential speed of the tracks and the thrust of the bow and stern horizontal thrusters are controlled by the track microcontroller to realize the multi-purpose unmanned transport platform turning underwater.

[0055] Furthermore, when the amphibious unmanned platform is in elevator mode, the horizontal rudder controls the vehicle's pitch angle, which, in conjunction with the propulsion of the two main thrusters, enables the amphibious unmanned platform to ascend or descend. Specifically, this includes...

[0056] Using depth error as the input to a single-neuron adaptive PID controller, the output is a horizontal rudder control command; combined with line-of-sight navigation, the desired X-axis position is defined as 10m ahead of the multi-purpose unmanned platform's own X-axis position. d Orientation and position, desired depth position is Z d The displacement error equation is obtained as follows:

[0057] x e =xx d ,z e =zz d

[0058] The desired pitch angle is:

[0059] θ d =tan -1 ((zz d ) / (xx d )).

[0060] Compared with the prior art, the beneficial effects of the present invention are:

[0061] This invention discloses an amphibious unmanned platform and its motion control method. The amphibious unmanned platform includes an upper platform, an underwater propulsion device, and a land-based power unit, combining the advantages of underwater propulsion and tracked land-based mobility. The bow and stern horizontal thrusters work together to enable the amphibious unmanned transport platform to turn in place, significantly reducing the turning radius. The bow and stern vertical thrusters work together to adjust the pitch angle of the amphibious unmanned transport platform, facilitating rapid concealment. The tracks enhance the amphibious unmanned transport platform's underwater and land obstacle-crossing capabilities, adapting to various terrain environments. The amphibious unmanned platform also employs wireless control, autonomously maneuvering according to a predetermined program, reducing its limitations during movement. Furthermore, it boasts advantages such as light weight, long remote control distance, long operating time, high efficiency, compact structure, and convenient operation. This invention also employs a PID controller based on radial basis function network tuning. Utilizing the arbitrary nonlinear expression capability of the radial basis function network, optimal PID control can be achieved through learning the system performance. Attached Figure Description

[0062] Figure 1 This is a front view of an amphibious unmanned platform according to an embodiment of the present invention;

[0063] Figure 2 This is a top view of an amphibious unmanned platform according to an embodiment of the present invention;

[0064] Figure 3 This is a schematic diagram of the stern direction of an amphibious unmanned platform according to an embodiment of the present invention;

[0065] Figure 4 This is a structural diagram of the radial basis function neural network in an embodiment of the present invention;

[0066] Figure 5 This is a block diagram of a PID controller tuned based on a radial basis function neural network in an embodiment of the present invention;

[0067] Figure 6 This is a simulation result diagram of a PID controller tuned using a radial basis function neural network in an embodiment of the present invention;

[0068] Figure 7 The track map of the multi-purpose unmanned platform obtained using the traditional PID control algorithm;

[0069] Figure 8 This is a flight path map of a multi-purpose unmanned platform obtained using a depth control method that combines line-of-sight navigation.

[0070] Explanation of reference numerals in the attached figures:

[0071] 2. Tracked chassis; 3. Battery compartment; 4. Electronics compartment; 5. External frame; 6. Buoyancy block; 201. Drive wheel; 202. Load wheel; 203. Guide wheel; 204. Track; 501. Left main thruster; 502. Right main thruster; 503. Stern vertical thruster; 504. Stern horizontal thruster; 505. Bow vertical thruster; 506. Bow horizontal thruster; 508. Doppler sensor. Detailed Implementation

[0072] In the description of this invention, it should be noted that the terms used in the various embodiments, such as "upper," "lower," "front," "rear," "left," and "right," which indicate orientation, are only used to simplify the description of the positional relationships based on the accompanying drawings and do not mean that the components and devices referred to must be operated in accordance with the specific orientations and defined operations, methods, and structures in the specification. Such directional terms do not constitute a limitation of this invention.

[0073] In the description of this invention, it should be noted that the terms "first," "second," and "third" mentioned in the embodiments of this invention are for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Therefore, a feature defined with "first," "second," and "third" may explicitly or implicitly include one or more of that feature.

[0074] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0075] Specific Implementation Plan 1: Combining Figures 1 to 3 As shown, the present invention provides an amphibious unmanned platform, including an upper platform, an underwater propulsion device, and a land-based power unit.

[0076] The upper platform includes an enclosed electronic cabin 4 and an external frame 5 outside the electronic cabin 4. The electronic cabin 4 is located at the center of the external frame 5. The electronic cabin 4 is equipped with a controlled system, a navigation and positioning system, a communication system and a power control system. The external frame 5 is equipped with an underwater propulsion system. The upper platform is equipped with a land-based power unit below it.

[0077] The underwater propulsion system includes a left main thruster 501 and a right main thruster 502 symmetrically arranged on both sides of the central axis of the stern of the amphibious unmanned platform, and also includes a stern vertical thruster 503, a stern horizontal thruster 504, a bow vertical thruster 505, and a bow horizontal thruster 506. The propellers of the left main thruster 501 and the right main thruster 502 are horizontally oriented towards the rear of the amphibious unmanned platform; the stern vertical thruster 503 is located at the bottom of the outer frame 5, and the propeller of the stern vertical thruster 503 is oriented towards the tracked chassis 2 of the land-based power unit; the stern horizontal thruster 504 and... The bow horizontal thrusters 506 are parallel to each other and located on the bow and stern sides of the amphibious unmanned platform respectively; the propellers of the bow vertical thrusters 505 are oriented perpendicular to the port side; the propellers of the bow horizontal thrusters 506 are oriented perpendicular to the starboard side; the coordinated use of the six thrusters enables the amphibious unmanned platform to perform maneuvers such as surfacing, diving, and bow turning; the thrusters are powered and controlled by the electronics compartment 4; the thrusters are powered by 48V watertight brushless DC motors that can switch between forward and reverse rotation at any time, adopt closed-loop control, provide real-time thrust data feedback, and have an overcurrent protection mechanism;

[0078] The land-based power system includes a tracked chassis 2 and two tracked assemblies located below the tracked chassis 2. Each tracked assembly includes a drive wheel 201, a guide wheel 203, a load wheel 202, a track 204, and a track motor. Several load wheels 202 are evenly distributed between the drive wheel 201 and the guide wheel 203, with the drive wheel 201 and the guide wheel 203 located diagonally above the load wheels 202 at both ends. The track 204 wraps around the drive wheel 201, the guide wheel 203, and the load wheels 202. The output end of the track motor is connected to the drive wheel 201. The drive wheel 201, the guide wheel 203, and the load wheels 202 are all rotatably connected to the tracked chassis 2. The amphibious unmanned platform achieves forward and backward movement and left and right turns on land through the forward and backward rotation and differential speed of the two tracked assemblies.

[0079] The controlled system includes a motion computer, a mission computer, and a guidance computer. The motion computer is used for multi-sensor data processing and for calculating pose control commands for the amphibious unmanned platform. The mission computer is used for mission planning, network communication, and fault monitoring. The guidance computer is used to provide guidance commands, including desired heading, desired speed, and desired depth. The motion control algorithm of the amphibious unmanned platform is run by the controlled system and can transmit the generated commands to the underwater propulsion system or the land-based power system, thereby controlling the movement of the amphibious unmanned platform in water, on land, or underwater.

[0080] The navigation and positioning system includes a GPS module inside the electronic cabin 4, an inertial integrated sensor module (including an inertial navigation sensor and a Doppler sensor 508 (DVL) located outside the cabin), and a depth gauge; it is used to perform positioning and attitude calculation for the amphibious unmanned platform, and achieves integrated navigation through the fusion of multiple sensors;

[0081] The communication system includes a WiFi communication module and a wireless communication module, which are used to connect to the host computer through the communication system. The communication method can be adjusted according to the environment. The hybrid communication method can ensure reliable communication of the multi-purpose unmanned transport platform over a wide area.

[0082] The power control system includes a main microcontroller control board and a relay matrix board. Each pin of the main microcontroller control board controls a relay switch, thereby controlling the power supply to each module. The relay board has a voltage divider module that divides the 48V battery voltage into three lines: 48V main power, 24V auxiliary power, and 5-12V control power, and controls them hierarchically through relays. The rated capacity of the battery is 300Wh, and the total battery capacity is 600Wh. The 48V main power is responsible for powering the six thrusters, the 24V auxiliary power is responsible for powering the high-power sensors and computer, and the 5-12V control power is responsible for powering the low-power sensors and microcontroller. The relays and current are sampled and processed by the microcontroller board and then sent to the mission computer. The electronics compartment 4 uses a dual power supply to avoid the current reduction caused by the operation of multiple motors, which would affect the overall performance of the multi-purpose unmanned platform. The power source is a rechargeable high-energy battery, which is located in the battery compartment 3 located at the stern of the external frame 5. The power control system is responsible for power management of all power devices and monitoring battery voltage and capacity.

[0083] Preferably, a plurality of buoyancy blocks 6 are provided above the tracked chassis 2. The buoyancy blocks 6 are made of foam-type buoyancy material. The multi-amphibious unmanned platform can be balanced by the buoyancy blocks 6. After balancing, the center of gravity and the center of buoyancy of the multi-amphibious unmanned platform are located at the geometric center. The overall balance is slightly negative buoyancy, which is conducive to the multi-amphibious unmanned platform sitting on the bottom and driving.

[0084] Preferably, the track motor can be powered by a battery on the track chassis 2. The track chassis 2 is equipped with a watertight control compartment. The control compartment contains a track microcontroller that directly controls the track motor and communicates with the controlled system via a wired network. The chassis control compartment is equipped with a water-cooling circulation system to cool the control compartment. The track motor is also equipped with an encoder to measure the movement speed of the track 204 by calculating the motor speed over a certain period of time.

[0085] Preferably, the electronic compartment 4 is a cylindrical stainless steel shell structure, which can effectively conduct heat in the underwater environment. The electronic compartment 4 has a heat dissipation module inside. Both ends of the shell of the electronic compartment 4 are provided with end caps equipped with watertight penetration parts. The end caps have rubber watertight interfaces, which can establish electrical and signal connections with external thrusters, sensors, and chassis components through watertight cables. The electronic compartment 4 ensures internal sealing through sealing rings and vacuum treatment.

[0086] Preferably, for the controlled system, a PID controller tuned based on radial basis function network is set. Radial basis function network is synonymous with RBF radial basis function network, RBF network and RBF neural network. PID adaptive control based on RBF radial basis function network is an effective control method. It combines the simplicity of PID controller and the nonlinear approximation capability of RBF network, and can better adapt to the control requirements of various complex systems.

[0087] Traditional PID control requires constant parameter adjustment, and inappropriate parameters make it difficult to achieve optimal results. Reaching the best performance can take a long time. Therefore, based on single-neuron adaptive PID control, a PID controller tuned using an RBF neural network is used. Compared to a single-neuron adaptive PID controller, its advantage lies in its nonlinearity. The control effect of a PID controller depends on the proportional, integral, and derivative control actions, forming a relationship of mutual cooperation and constraint among the control quantities. However, the effect is not necessarily linear, so using a nonlinear RBF can achieve better results. The arbitrary nonlinear expression capability of the RBF neural network allows for the achievement of optimal PID control combinations through learning system performance.

[0088] Radial basis function neural networks (RBNs) are three-layer feedforward networks with a single hidden layer. The input-to-output transition is non-linear, while the mapping from the hidden layer space to the output is linear, accelerating the learning process and avoiding local minima. Their structure is as follows: Figure 4 As shown, in the RBF network structure, X = [x1, x2, ..., x...]. n ] T Let H = [h1, h2, ..., h2] be the radial basis vector of the hidden layer nodes in the RBF neural network, and let H be the input vector of the network. m ] T , where h j For Gaussian functions:

[0089]

[0090] The center vector of the j-th node in the hidden layer of the RBF neural network is C. j =[c j1 ,c j2 ,...,c jm ]T Let the basis width vector of the hidden layer nodes in the RBF neural network be B = [b1, b2, ..., b m ] T , where b j Let w be the base width parameter of hidden layer node j, and it is a number greater than 0; the weight vector of the network is w = [w1, w2, ..., w...]. n ] T The network output for identification is:

[0091] y mout =w1h1+w2h2+...+w m h m

[0092] Combination Figure 5 The diagram shown is a block diagram of a PID controller based on RBF network tuning. The parameter tuning algorithm in the network is NNI parameter tuning. Let the theoretical output of the identified system at time k be y. out (k), the output of the network identification is y mout (k), then the system performance index function is:

[0093]

[0094] The iterative algorithm for obtaining the output weights, node centers, and node base width parameters using gradient descent is as follows:

[0095] The output weight expression is as follows:

[0096] w j (k)=w j (k-1)+η(y out (k)-y mout (k))h j +α(w j (k-1)-w j (k-2))+β(w j (k-2)-w j (k-3))

[0097] Where η is the learning rate; α and β are both momentum factors, with values ​​ranging from (0, 1);

[0098] Node base width variation Δb j for:

[0099]

[0100] Node base width parameter b j for:

[0101] b j (k)=b j (k-1)+ηΔbj +α(b j (k-1)-b j (k-2))+β(b j (k-2)-b j (k-3))

[0102] Node center change Δc ji for:

[0103]

[0104] Node center c j for:

[0105] c ji =c ji (k-1)+ηΔc ji +α(c ji (k-1)-c ji (k-2))+β(c ji (k-2)-c ji (k-3))

[0106] Where cj = [c11, c12, c13…c1m] is the center vector value of the j-th hidden layer neuron;

[0107] The PID parameter tuning algorithm is as follows:

[0108] The control error is:

[0109] error(k) = u(k) - yout(k)

[0110] Where u(k) is the input of the controlled system and yout(k) is the output of the controlled system;

[0111] The three PID inputs are:

[0112] xc(1) = error(k) - error(k-1)

[0113] xc(2) = error(k)

[0114] xc(3)=error(k)-2error(k-1)+error(k-2)

[0115] The neural network tuning index is:

[0116]

[0117] The three parameters K are obtained by gradient descent. p ,K I ,K D The adjustment amount is:

[0118]

[0119]

[0120]

[0121] in, Jacobian information of the controlled object, i.e., the sensitivity of the object's output to changes in the control input, can be identified using an RBF neural network; K p For the proportional gain term, K i K is the integral gain term. d This is the differential gain term;

[0122]

[0123] Using the desired depth and the actual depth of the multi-purpose unmanned platform as errors, the RBF network structure is 3-6-1. The three parameters for network identification are: control input increment, current depth, and previous depth information. The initial proportional, integral, and derivative parameters are set to 0.01, 0.02, and 0.03, respectively. When the simulation ends, these parameters become 436.5727, 60.9628, and 1117, showing that the three parameters are continuously adjusted during the simulation. The simulation results are as follows: Figure 6 As shown;

[0124] Specific Implementation Plan Two: Combining Figures 1 to 3 As shown, the present invention provides a motion control method for an amphibious unmanned platform, comprising the following steps:

[0125] S100. When the amphibious unmanned platform moves in water, it is driven only by the propulsion system. The motion is decomposed into horizontal and vertical motion using underactuated control. The thrust of the bow vertical thruster 505 and stern vertical thruster 503 is controlled by the controlled system to keep the amphibious unmanned platform at a constant depth. The thrust of the bow horizontal thruster 506 and stern horizontal thruster 504 is controlled by the controlled system to control the bow turning and navigation of the amphibious unmanned platform. The forward speed of the amphibious unmanned platform is changed by controlling the thrust output of the thrusters.

[0126] When the amphibious unmanned platform is in elevator mode, to achieve ascent or descent, the only method is to control the vehicle's pitch angle using the horizontal rudder, in conjunction with the two main thrusters, to achieve ascent or descent, i.e., depth control; specifically including...

[0127] Combination Figure 7The diagram shows a traditional PID control algorithm, using depth error as the input to the PID controller and outputting a horizontal rudder control command. It can be seen that when using depth error as the controller input, oscillations always occur when adjusting parameters. Therefore, a line-of-sight navigation method is used, with the desired X-axis position 10m ahead of the multi-purpose unmanned platform as the target position. d The orientation and position, and the desired depth position are Z. d Therefore, the displacement error equation can be obtained:

[0128] x e =xx d ,z e =zz d

[0129] Based on the line-of-sight navigation method, the desired pitch angle can be obtained, and its expression is as follows:

[0130] θ d =tan -1 ((zz d ) / (xx d ))

[0131] Combination Figure 8 As shown, the desired depths are 100m and 60m respectively. As can be seen from the navigation trajectory diagram of the multi-purpose unmanned platform, the depth control method combined with the line-of-sight navigation method can achieve better results. It can enable the multi-purpose unmanned platform to converge to the specified position, thereby solving the depth control problem. However, the choice of the specified position will have a significant impact on the final control effect. If the position in the X direction is too far, it will slow down the convergence to the desired depth. If it is too small, it will fail to converge to the desired depth.

[0132] S200: When the multi-purpose unmanned platform moves on land, it is driven only by the land-based power unit. The differential speed of the track is controlled by the track microcontroller in the control cabin set on the tracked chassis 2 to realize the turning of the multi-purpose unmanned platform on land. The speed of the track 204 is controlled by controlling the duty cycle of the track motor to realize the maneuvering mode of acceleration, deceleration, forward and backward.

[0133] S300: When the multi-purpose unmanned platform moves underwater, it uses a combination of propulsion system and land-based power unit to drive the multi-purpose unmanned platform to crawl along the seabed. By controlling the bow vertical thruster 505 and stern vertical thruster 503 to apply downward thrust, the pressure between the multi-purpose unmanned platform and the seabed is increased. The left main thruster 501 and right main thruster 502 are controlled to assist the multi-purpose unmanned platform in walking underwater. The differential speed of the track 204 and the thrust of the bow horizontal thruster 506 and stern horizontal thruster 504 are controlled by the track microcontroller to realize the multi-purpose unmanned transport platform turning underwater.

[0134] The other combinations and connections in this implementation scheme are the same as in Specific Implementation Scheme 1.

[0135] While the present invention has been disclosed above, its scope of protection is not limited thereto. Those skilled in the art can make various changes and modifications without departing from the spirit and scope of the present invention, and all such changes and modifications will fall within the scope of protection of the present invention.

Claims

1. An amphibious unmanned platform, characterized in that: Including the platform, underwater propulsion system, and land-based power system. The upper platform includes a closed electronic cabin (4) and an external frame (5) outside the electronic cabin (4). The electronic cabin (4) is located at the center of the external frame (5). The electronic cabin (4) is equipped with a controlled system, a navigation and positioning system, a communication system and a power control system. The external frame (5) is equipped with an underwater propulsion system. The upper platform is equipped with a land-based power system below it. The underwater propulsion system includes a left main thruster (501) and a right main thruster (502) symmetrically arranged on both sides of the central axis of the stern of the multi-purpose unmanned platform, and also includes a stern vertical thruster (503), a stern horizontal thruster (504), a bow vertical thruster (505), and a bow horizontal thruster (506). The propellers of the left main thruster (501) and the right main thruster (502) are horizontally oriented towards the rear of the multi-purpose unmanned platform; the stern vertical thruster (503)... 03) The propeller of the stern vertical thruster (503) is located at the bottom of the external frame (5) and faces the tracked chassis (2) of the land power unit; the stern horizontal thruster (504) and the bow horizontal thruster (506) are parallel to each other and are located on the bow and stern sides of the amphibious unmanned platform respectively; the propeller of the bow vertical thruster (505) faces the direction perpendicular to the port side; the propeller of the bow horizontal thruster (506) faces the direction perpendicular to the starboard side; The land-based power system includes a tracked chassis (2) and two track assemblies located below the tracked chassis (2). Each track assembly includes a drive wheel (201), a guide wheel (203), a load wheel (202), a track (204), and a track motor. Several load wheels (202) are evenly distributed between the drive wheel (201) and the guide wheel (203), and the drive wheel (201) and the guide wheel (203) are both located diagonally above the load wheels (202) at both ends. The track (204) wraps around the drive wheel (201), the guide wheel (203), and several load wheels (202). The output end of the track motor is connected to the drive wheel (201). The drive wheel (201), the guide wheel (203), and the load wheel (202) are all rotatably connected to the tracked chassis (2).

2. The amphibious unmanned platform according to claim 1, characterized in that: The motion computer is used for multi-sensor data processing and to calculate the pose control commands for the multi-amphibious unmanned platform; the mission computer is used to execute mission planning and to perform network communication and fault monitoring; and the guidance computer is used to provide guidance commands, including the desired heading, desired speed and desired depth.

3. The amphibious unmanned platform according to claim 2, characterized in that: The navigation and positioning system includes a GPS module and an inertial navigation sensor inside the electronic cabin (4), as well as a Doppler sensor (508) and a depth gauge located outside the electronic cabin; used for positioning and attitude calculation of the amphibious unmanned platform; the communication system includes a WiFi communication module and a wireless communication module, used for the amphibious unmanned platform to connect to the host computer through the communication system.

4. The amphibious unmanned platform according to claim 3, characterized in that: The power control system includes a main microcontroller control board and a relay matrix board. Each pin of the main microcontroller control board is used to control a relay switch and thus control the power supply of each module. The electronic cabin (4) is powered by a dual power supply, which is a battery.

5. The amphibious unmanned platform according to claim 4, characterized in that: Several buoyancy blocks (6) are provided above the tracked chassis (2).

6. The amphibious unmanned platform according to claim 5, characterized in that: The track motor is powered by a battery installed on the track chassis (2). The track chassis (2) is equipped with a watertight control cabin, which contains a track microcontroller. The track microcontroller is used to control the track motor and communicate with the controlled system via a wired network.

7. The amphibious unmanned platform according to claim 6, characterized in that: For the controlled system, a PID controller tuned based on an RBF neural network is set up, including, Let X = [x1, x2, ..., x n ] T Given the network input vector, the radial basis vectors of the hidden layer nodes are H = [h1, h2, ..., h...]. m ] T h j For Gaussian functions: The center vector of the j-th node in the hidden layer of the RBF neural network is C. j =[c j1 ,c j2 ,...,c jm ] T Let the basis width vector of the hidden layer nodes in the RBF neural network be B = [b1, b2, ..., b m ] T , where b j Let w be the base width parameter of hidden layer node j, and it is a number greater than 0; the weight vector of the network is w = [w1, w2, ..., w...]. n ] T The network output for identification is: y mout =w1h1+w2h2+...+w m h m Let the theoretical output of the system to be identified at time k be y. out (k), the output of the network identification is y mout (k), then the system performance index function is: The output weights, node centers, and node base width parameters are obtained using the gradient descent method. The output weight expression is as follows: w j (k)=w j (k−1)+η(y out (k)-y mout ( k )) h j +α(w j (k-1)-w j (k-2))+β(w j (k-2)-w j (k-3)) Where η is the learning rate; α and β are both momentum factors, with values ​​ranging from (0, 1); Node base width variation Δb j for: Node base width parameter b j for: b j (k)=b j (k-1)+ηΔb j +α(b j (k-1)-b j (k-2))+β(b j (k-2)-b j (k-3)) Node center change Δc ji for: Node center c j for: c ji =c ji (k-1)+ηΔc ji +α(c ji (k-1)-c ji (k-2))+β(c ji (k-2)-c ji (k-3)) Where cj = [c11, c12, c13, ..., c1m]; The PID parameter tuning algorithm is as follows: The control error is: error(k) = u(k) - yout(k) Where u(k) is the input of the controlled system and yout(k) is the output of the controlled system; The three PID inputs are: xc(1) = error(k) - error(k-1) xc(2) = error(k) xc(3)=error(k)-2error(k-1)+error(k-2) The neural network tuning index is: The three parameters K are obtained by gradient descent. p ,K I ,K D The adjustment amount in decibels is: in, The sensitivity of the object's output to changes in the control input; K p For the proportional gain term, K i K is the integral gain term. d This is the differential gain term; 8. A motion control method for an amphibious unmanned platform according to any one of claims 1-7, characterized in that, Includes the following steps: S100. When the multi-purpose unmanned platform moves in the water, it is driven only by the propulsion system. The motion is decomposed into horizontal and vertical motion by underactuated control. The multi-purpose unmanned platform is kept at a constant depth by controlling the thrust of the bow vertical thruster (505) and stern vertical thruster (503). The multi-purpose unmanned platform is turned and navigated by controlling the thrust of the bow horizontal thruster (506) and stern horizontal thruster (504). The forward speed of the multi-purpose unmanned platform is changed by controlling the thrust output of the thrusters. S200. When the multi-purpose unmanned platform moves on land, it is driven by only the land power unit. The differential speed of the track (204) is controlled by the track single-chip microcomputer in the control cabin set on the track chassis (2) to realize the turning of the multi-purpose unmanned platform on land. The speed of the track (204) is controlled by controlling the duty cycle of the track motor to realize the acceleration, deceleration, forward and backward movement modes. S300. When the multi-purpose unmanned platform moves underwater, it is driven by a combination of propulsion system and land power unit. The pressure between the multi-purpose unmanned platform and the seabed is increased by controlling the downward thrust of the bow vertical thruster (505) and the stern vertical thruster (503). The left main thruster (501) and the right main thruster (502) are controlled to assist the multi-purpose unmanned platform in walking underwater. The differential speed of the track (204) and the thrust of the bow horizontal thruster (506) and the stern horizontal thruster (504) are controlled by the track microcontroller to realize the turning of the multi-purpose unmanned transport platform underwater.

9. The motion control method for an amphibious unmanned platform according to claim 8, characterized in that: When the amphibious unmanned platform is in elevator mode, the horizontal rudder controls the vehicle's pitch angle, which, in conjunction with the propulsion of the two main thrusters, enables the platform to ascend or descend. Specifically, this includes... Using depth error as the input to a single-neuron adaptive PID controller, the output is a horizontal rudder control command; combined with line-of-sight navigation, the desired X-axis position is defined as 10m ahead of the multi-purpose unmanned platform's own X-axis position. d Orientation and position, desired depth position is Z d The displacement error equation is obtained as follows: x e =x-x d ,z e =z-z d The desired pitch angle is: θ d <tan -1 ((zz d ) / (xx d ))。

Images