State-augmented air-ground amphibious vehicle path planning and mode switching method and system

Abstract

The invention discloses a state-augmented air-ground amphibious vehicle path planning and mode switching method and system, and relates to the field of unmanned vehicles, and the method comprises the steps: constructing a dynamic model of an air-ground amphibious vehicle; constructing a dynamic boundary constraint condition of the four-duct flight unit; acquiring original data of a running path of the land-air amphibious vehicle; generating a shortest path of the air-ground amphibious vehicle by adopting an A * algorithm based on the original data, the dynamic model of the air-ground amphibious vehicle and the dynamic boundary constraint condition; and optimizing the shortest path based on the state augmentation amount to obtain an optimized path of the land-air amphibious vehicle, and completing the path planning of the land-air amphibious vehicle. The state augmentation quantity represents a driving mode of the land-air amphibious vehicle; and switching the running modes of the land-air amphibious vehicle based on the optimized path. The safety and the driving efficiency of the land-air amphibious vehicle in the driving process are improved.

Description

Technical Field

[0001] The present application relates to the field of unmanned vehicles, and in particular to a state-augmented amphibious vehicle path planning and mode switching method and system. Background Art

[0002] Amphibious vehicles combine the rapid maneuverability of aerial drones with the long-range endurance of ground vehicles, attracting significant attention in complex air-to-ground scenarios such as post-disaster emergency rescue, underground facility inspections, and convenient low-altitude travel. As a new type of unmanned equipment, amphibious vehicles integrate unmanned operation and intelligent technologies. Their systems consist of multiple collaborative modules, including environmental perception, path planning, and motion control. The path planning module plays a crucial role in connecting the upper and lower levels. Its core task is to design a collision-free, safe path from the starting point to the destination. This path primarily involves ground travel, while also incorporating a collision-free, safe path from the starting point to the destination through aerial flight. The design also achieves comprehensive optimization of path time, distance, and energy consumption, posing significant challenges to current research on intelligent amphibious systems.

[0003] However, at present, amphibious path planning algorithms all use vertical height values ​​as the judgment conditions for land-air switching. When amphibious vehicles perform tasks in complex environments, such as unstructured complex scenes such as slopes and gravel, when amphibious vehicles work in an environment with gradually changing heights such as slopes, the height value at this time cannot be consistent with the passability of the environment. As the amphibious platform moves, when the height exceeds the set threshold, the platform planning algorithm will consider that the mode needs to be switched, that is, it cannot take off during the climbing process, or cannot avoid negative obstacles such as potholes in time, resulting in low safety. At the same time, even if avoidance is performed, the amphibious path planning algorithm needs to perform complex calculations, which in turn reduces the driving efficiency of the amphibious vehicle. Summary of the Invention

[0004] The purpose of this application is to provide a state-augmented amphibious vehicle path planning and mode switching method and system, which can improve the safety and driving efficiency of amphibious vehicles during driving.

[0005] To achieve the above objectives, this application provides the following solutions.

[0006] In a first aspect, the present application provides a state-augmented amphibious vehicle path planning and mode switching method, which includes the following contents.

[0007] A dynamic model of an amphibious vehicle is constructed; the amphibious vehicle includes a four-ducted flight unit and a tracked travel unit.

[0008] Construct the dynamic boundary constraints of the four-ducted flight unit.

[0009] The original data of the amphibious vehicle's driving path is obtained; the original data includes: starting coordinates and end point coordinates.

[0010] Based on the original data, the dynamic model of the amphibious vehicle and the dynamic boundary constraints, an A* algorithm is used to generate the shortest path for the amphibious vehicle to travel.

[0011] The shortest path is optimized based on the state augmentation amount to obtain the optimized path for the amphibious vehicle to travel, and the path planning of the amphibious vehicle is completed; the state augmentation amount represents the travel mode of the amphibious vehicle.

[0012] The driving mode of the amphibious vehicle is switched based on the optimized path.

[0013] Optionally, the dynamic model of the amphibious vehicle includes a system model of a four-ducted flight unit and a dynamic model of a tracked travel unit.

[0014] The expression of the system model of the four-ducted flight unit is as follows.

[0015]

[0016] in, is the acceleration of the amphibious vehicle in the x-axis direction; is the acceleration of the amphibious vehicle in the y-axis direction; is the acceleration of the amphibious vehicle in the z-axis direction; g is the acceleration due to gravity; u t is the time input; u φ is the roll direction input; u θ is the yaw direction input; u ψ is the pitch direction input; m is the total mass of the amphibious vehicle; φ is the roll angle of the amphibious vehicle; θ is the yaw angle of the amphibious vehicle; ψ is the pitch angle of the amphibious vehicle; J=diag{J xx ,J yy ,J zz} is the moment of inertia of the amphibious vehicle relative to the inertial coordinate axis; is the angular acceleration of the amphibious vehicle around the x-axis; is the angular acceleration of the amphibious vehicle around the y-axis; is the angular acceleration of the amphibious vehicle around the z axis; r is the pitch angular rate of the amphibious vehicle; p is the roll angular rate of the amphibious vehicle; q is the yaw angular rate of the amphibious vehicle; J r is the moment of inertia of the motor; w represents the speed difference between the i-th motors with opposite rotation directions.

[0017] The expression of the dynamic model of the crawler traveling unit is as follows.

[0018]

[0019] Among them, T l is the output torque of the left motor of the amphibious vehicle; T r is the output torque of the right motor of the amphibious vehicle; intermediate variable β=sinα+fcosα, α is the slope of the ramp, f is the rolling resistance coefficient; r l is the driving wheel rotation radius of the amphibious vehicle; δ is the rotation mass conversion coefficient; The speed of the amphibious vehicle.

[0020] Optionally, the dynamic boundary constraints of the four-ducted flight unit are expressed as follows.

[0021] f p =λ dc f dc +λ dv f dv +λ da f da .

[0022] Among them, λ dc is the collision weight coefficient; λ dv is the weight coefficient of the speed dynamics feasibility constraint; da is the acceleration dynamics feasibility constraint weight coefficient; f dc is the collision cost between the ducted propeller structure and the obstacle; f dv is the rotation speed penalty of the ducted propeller, f da is the rotational acceleration cost of the ducted propeller.

[0023] Optionally, based on the original data, the dynamic model of the amphibious vehicle and the dynamic boundary constraints, an A* algorithm is used to generate the shortest path for the amphibious vehicle to travel, which specifically includes the following contents.

[0024] Based on the original data, the dynamic model of the amphibious vehicle and the dynamic boundary constraints, an A* algorithm is used to generate initial control points of three-dimensional state quantities.

[0025] The shortest path for the amphibious vehicle to travel is determined based on the initial control point.

[0026] Optionally, the shortest path is optimized based on the state augmentation amount to obtain an optimized path for the amphibious vehicle, which specifically includes the following contents.

[0027] Based on the shortest path, the cost of the amphibious vehicle is calculated.

[0028] The state augmentation cost is calculated by weighted summing of the land-air amphibious vehicle costs, and a state augmentation objective function is constructed.

[0029] Based on the state augmentation objective function, the initial control points are transformed into control points of the four-dimensional state quantity to obtain the transformed control points.

[0030] Based on the transformed control points, the elastic band principle is used to calculate the cost of amphibious vehicles.

[0031] The total kinematic cost of the amphibious vehicle is calculated by taking a weighted sum of the amphibious vehicle costs.

[0032] Based on the total kinematic cost of the amphibious vehicle, the shortest path of the amphibious vehicle is optimized to generate the optimized path of the amphibious vehicle.

[0033] Optionally, the cost of the amphibious vehicle includes: distance cost, energy consumption cost and mode switching cost.

[0034] The distance cost is calculated as follows.

[0035]

[0036] Among them, F d is the distance cost of amphibious vehicles; l j is the length of the jth line segment in the shortest path, where the line segment is the distance between two adjacent initial control points.

[0037] The formula for calculating energy consumption cost is as follows.

[0038] F e =ω land vl land +ω air ∑l air .

[0039] Among them, F e is the energy consumption cost of the amphibious vehicle; ω land is the cost of the amphibious vehicle when traveling on the ground; ∑l land is the sum of the lengths of the multi-line segments of the amphibious vehicle when it is traveling on the ground; ω air is the cost of the amphibious vehicle in flight; ∑l air It is the sum of the lengths of the polyline segments of the amphibious vehicle when it is flying.

[0040] The calculation formula of mode switching cost is as follows.

[0041] F s =N sup C sup +Nsdown C sdown .

[0042] Among them, F s is the mode switching cost of the amphibious vehicle; N sup The number of times ground travel is converted into flight travel; C sup Cost of converting ground travel to flight travel; N sdown The number of ground trips converted from flight trips; C sdown The cost of converting ground travel to flight travel.

[0043] Optionally, the calculation formula of the state augmentation cost is as follows.

[0044] F=ω1F d +ω2F e +ω3F s .

[0045] The expression of the state augmentation objective function is as follows.

[0046]

[0047] Among them, W is the state augmentation objective function; 1 is the amphibious vehicle in the flight state; 0 is the amphibious vehicle in the ground driving state; F is the state augmentation cost; F T The state augmentation threshold.

[0048] Optionally, the amphibious vehicle cost includes: smoothness cost, collision cost, speed feasibility cost and acceleration feasibility cost.

[0049] The calculation formula of the smoothness cost is as follows.

[0050]

[0051] Among them, f s is the smoothness cost of the amphibious vehicle; N is the total number of initial control points; p b is the order of the B-spline curve; Q i+1 is the control point after the i+1th transformation; Q i is the control point after the i-th transformation; Q i-1 is the control point after the i+1th transformation.

[0052] The calculation formula of the collision cost is as follows.

[0053]

[0054] Among them, f c is the collision cost of amphibious vehicles; d(Q i ) is the control point Q after transformationi Distance to obstacles; d thr The distance between the amphibious vehicle and the obstacle.

[0055] The calculation formula of speed feasibility cost is as follows.

[0056]

[0057] Among them, f v is the speed feasibility cost of the amphibious vehicle; μ is the {x, y, z} direction of the control point of the amphibious vehicle after the transformation; The control point Q of the amphibious vehicle after the transformation i Speed; v max The maximum speed of an amphibious vehicle.

[0058] The calculation formula of acceleration feasibility cost is as follows.

[0059]

[0060] Among them, f a The acceleration feasibility cost of amphibious vehicles; The control point Q of the amphibious vehicle after the transformation i acceleration; a max It is the maximum acceleration of an amphibious vehicle.

[0061] Alternatively, the total kinematic cost of the amphibious vehicle is calculated as follows.

[0062] f1=λ s f s +λ c f c +λ f (f v +f a ).

[0063] Where f1 is the total kinematic cost of the amphibious vehicle; s ,λ c ,λ f are weight values; f s is the smoothness cost of amphibious vehicles; f c is the collision cost of amphibious vehicles; f v is the speed feasibility cost of the amphibious vehicle; f a The acceleration feasibility cost of amphibious vehicles.

[0064] In the second aspect, the present application provides a state-augmented amphibious vehicle path planning and mode switching system, which includes the following modules.

[0065] The dynamic model construction module is used to construct a dynamic model of an amphibious vehicle; the amphibious vehicle includes a four-ducted flight unit and a tracked travel unit.

[0066] The dynamic boundary constraint condition construction module is used to construct the dynamic boundary constraint conditions of the four-ducted flight unit.

[0067] The acquisition module is used to obtain the original data of the amphibious vehicle's driving path; the original data includes: starting coordinates and end point coordinates.

[0068] The shortest path generation module is used to generate the shortest path for the amphibious vehicle to travel by using the A* algorithm based on the original data, the dynamic model of the amphibious vehicle and the dynamic boundary constraints.

[0069] The optimized path generation module is used to optimize the shortest path based on the state augmentation amount to obtain the optimized path for the amphibious vehicle and complete the path planning of the amphibious vehicle; the state augmentation amount represents the driving mode of the amphibious vehicle.

[0070] A driving mode switching module is used to switch the driving mode of the amphibious vehicle based on the optimized path.

[0071] According to the specific embodiments provided in this application, this application has the following technical effects.

[0072] The present application provides a state-augmented amphibious vehicle path planning and mode switching method and system. During the driving process of the amphibious vehicle, the A* algorithm is used to generate the shortest path of the amphibious vehicle through original data, the dynamic model of the amphibious vehicle and the dynamic boundary constraints. The shortest path is further optimized by the state augmentation amount (the driving mode of the amphibious vehicle) to obtain the optimized path of the amphibious vehicle, completing the path planning of the amphibious vehicle. The amphibious unmanned vehicle not only has coordinates but also the driving mode of the amphibious vehicle during the path driving process, so that the amphibious vehicle can switch the driving state in advance and predict the driving state of the amphibious vehicle at the next moment, so that the amphibious vehicle can change the driving state in advance and quickly complete the vehicle driving state change, thereby improving the safety and driving efficiency of the amphibious vehicle during the driving process. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0074] Figure 1 A schematic diagram of an amphibious vehicle provided in one embodiment of the present application, wherein (a) is a front view of the amphibious vehicle, and (b) is a right front view of the amphibious vehicle.

[0075] Figure 2 A schematic diagram of the structure of an amphibious vehicle provided in one embodiment of the present application.

[0076] Figure 3 A flow chart of a state-augmented amphibious vehicle path planning and mode switching method provided for an implementation of the present application.

[0077] Figure 4 A schematic diagram of the dynamic structure of a ducted flight unit provided in one embodiment of the present application.

[0078] Figure 5 Schematic diagram of a simplified dynamic model of a crawler chassis provided in one embodiment of the present application.

[0079] Figure 6 Schematic diagram of three environmental effects produced by aerodynamic characteristics provided in one embodiment of the present application.

[0080] Figure 7 A schematic diagram of the calculation process of the state augmentation amount provided in one embodiment of the present application.

[0081] Figure 8 A schematic diagram of the calculation process of the state augmentation amount provided in one embodiment of the present application.

[0082] Figure 9 A schematic diagram of a vehicle sensor and data communication architecture provided in one embodiment of the present application.

[0083] Figure 10 Schematic diagram of testing of an amphibious vehicle in different environments provided in one embodiment of the present application, wherein (a) is a test of passability in a ground obstacle environment, (b) is a test of maneuverability in an air obstacle environment, and (c) is a test of high maneuverability in a three-dimensional narrow space environment.

[0084] Figure 11 Schematic diagram of the simulation effect of the optimal path under different environments provided by one embodiment of the present application, wherein (a) is the simulation effect of the optimal path in a ground obstacle crossing environment, (b) is the simulation effect of the optimal path in an aerial obstacle crossing environment, and (c) is the simulation effect of the optimal path in a three-dimensional narrow space environment.

[0085] Figure 12Schematic diagrams of path trajectories and state augmentation trajectories for different methods in different environments provided by one embodiment of the present application, where (a) is the trajectory in a ground obstacle crossing environment, (b) is the trajectory in an aerial obstacle crossing environment, and (c) is the trajectory in a three-dimensional narrow space environment.

[0086] Figure 13 Schematic diagram of the change of movement speed under different methods in different environments provided by one embodiment of the present application, where (a) is the movement speed in a ground obstacle crossing environment, (b) is the movement speed in an aerial obstacle crossing environment, and (c) is the movement speed in a three-dimensional narrow space environment.

[0087] Reference numerals: 1-camera, 2-onboard computer, 3-vehicle controller, 4-communication antenna, 5-battery, 6-drive wheel, 7-chassis suspension, 8-nylon track, 9-propeller, 10-receiver, 11-carbon fiber shell. DETAILED DESCRIPTION

[0088] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0089] There are currently several related studies on path planning methods for amphibious vehicles. Patent publication number CN116400703A proposes a path planning method, apparatus, computer equipment, and storage medium. This method uses an improved kinematics A* algorithm to generate an initial path, distinguishes the speed and acceleration of amphibious equipment in air and ground modes, switches the modes based on the altitude of the path points, and finally optimizes the initial path using a piecewise B-spline curve. Patent publication number CN118565474A proposes a combined air and land motion planning method for a tensegrity cross-domain robot. This method randomly samples the feasible space and determines the visual relationships between sampled points and connecting paths to obtain the topological information and guidance path of the feasible space. The guidance path is then fed into the air trajectory planning algorithm and the surface trajectory planning algorithm, respectively, to obtain the air trajectory and surface path. The invention patent with application number and publication number CN117784817A proposes an integrated planning, control and method for land and air amphibious unmanned platforms. It uses a three-dimensional grid map generated in real time to plan the safe operation trajectory of the land and air amphibious unmanned platform to avoid obstacles in real time, and determines the vertical height of the land and air safe operation trajectory to determine the movement mode of the land and air amphibious unmanned platform.

[0090] However, the current research on path planning methods uses vertical height values ​​as the conditions for judging the switching of land-air planning. When the land-air amphibious unmanned platform performs tasks in a complex environment, it is difficult to maintain a relatively flat driving environment. When the land-air amphibious platform works in an environment with gradually changing heights such as a slope, the height value at this time cannot be consistent with the passability of the environment. As the land-air amphibious platform moves, when the height exceeds the set threshold, the platform planning algorithm will consider that the mode needs to be switched, that is, it will not be able to take off during the climbing process, or it will not be able to avoid negative obstacles such as pits in time.

[0091] In the research on land-air path planning algorithms, the invention patent application number and publication number CN117687435A proposes a path planning method, device, electronic device, and storage medium for land-air amphibious drones. The method calculates the shading intensity of the land-air amphibious drone in the path planning grid. Based on the shading intensity and light density of the land-air amphibious drone in the path planning grid, speed and acceleration ranges based on the plant growth algorithm are established to generate the planned path for the land-air amphibious drone. The invention patent application number and publication number CN117032274A proposes an AGA-UAV path planning method based on an improved constrained multi-objective optimization evolutionary algorithm. Based on the improved constrained multi-objective optimization evolutionary algorithm, a constrained multi-objective optimization evolutionary model for land-air amphibious drone path planning is constructed. The optimal land-air amphibious drone path planning path is obtained by executing the constrained multi-objective optimization evolutionary algorithm. The invention patent with application number and publication number CN117057156A proposes a path search method for land and air amphibious aircraft. By initializing the grid map, the real environment is discretized into a grid map, the environmental information is recorded in the grid, the path search start and end points are set, and the path is backtracked by querying the node with the lowest comprehensive cost to generate a land and air trajectory.

[0092] However, the current research on land-air path planning algorithms lacks a method that can obtain the land-air amphibious vehicle mode state information in advance based on environmental information and land-air amphibious vehicle state information, and then add the land-air amphibious vehicle mode state as an information quantity into the design and research of the path planning algorithm function.

[0093] To address these shortcomings, the time it takes to switch between land and air modes for amphibious vehicles is shortened, improving their efficiency. Simultaneously, trajectory generation takes into account the platform's dynamic constraints and the aerodynamic performance of the duct structure, making the amphibious vehicle more stable and safe during movement.

[0094] The present application aims to provide a state-augmented path planning and mode switching method for amphibious vehicles. During the path planning process of amphibious vehicles, the amphibious vehicles can not only obtain general state information such as the position and speed of trajectory points, but also obtain the modes of the amphibious vehicles at different positions. This can help the amphibious vehicles understand their state changes in the future in advance, formulate strategies in advance, and pass through complex environments that require switching between land and air. This method can greatly improve the environmental adaptability and passability of amphibious vehicles in non-mechanical environments, improve the safety of amphibious vehicles during driving, and at the same time shorten the time for amphibious vehicles to switch between land and air modes, thereby improving the driving efficiency of amphibious vehicles.

[0095] The amphibious vehicle that matches this application uses a combination of tracks and ducts to achieve land and air maneuvers, and has the ability to be remotely controlled or intelligently autonomously fly in complex environments. The design of the amphibious vehicle mainly includes five parts: ground driving system, air flight system, perception and positioning system, flight control part, and power supply part. In order to ensure the performance of the state augmentation planning algorithm, the overall design of the amphibious vehicle is mainly targeted at the ground driving system, air flight system, perception and positioning system. The air flight unit of the amphibious vehicle adopts an X-shaped layout, the onboard computing and control system is set in the center of the body, and the flight power system is 4 duct modules arranged symmetrically relative to the body. At the same time, independent electric drive crawler driving chassis are installed on the left and right sides of the platform. The overall size of the model is 60cm long × 60cm wide. The overall design weight of the amphibious vehicle is 4.3kg, and the overall structure is very compact. The main view and right front view of the amphibious vehicle are as shown below. Figure 1 shown.

[0096] The specific structure of the land and air amphibious vehicle is as follows Figure 2 As shown, it includes: a camera 1, an onboard computer 2, a vehicle controller 3, a communication antenna 4, a battery 5, a drive wheel 6, a chassis suspension 7, a nylon track 8, a propeller 9, a receiver 10, and a carbon fiber shell 11.

[0097] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0098] In a specific embodiment, Figure 3 As shown, a state-augmented amphibious vehicle path planning and mode switching method is provided, which includes the following steps S1 to S6.

[0099] Step S1: Construct a dynamic model of an amphibious vehicle; the amphibious vehicle includes a four-ducted flight unit and a tracked travel unit.

[0100] Step S2: Construct the dynamic boundary constraints of the four-ducted flight unit.

[0101] Step S3: obtaining the original data of the amphibious vehicle's travel path; the original data includes: starting coordinates and end point coordinates.

[0102] Step S4: Based on the original data, the dynamic model of the amphibious vehicle and the dynamic boundary constraints, the A* algorithm is used to generate the shortest path for the amphibious vehicle to travel.

[0103] Step S5: Optimizing the shortest path based on the state augmentation amount to obtain the optimized path for the amphibious vehicle, thereby completing the path planning for the amphibious vehicle; the state augmentation amount represents the driving mode of the amphibious vehicle.

[0104] Step S6: Switching the driving mode of the amphibious vehicle based on the optimized path.

[0105] In a specific embodiment, the dynamic model of the amphibious vehicle in step S1 includes a system model of a four-ducted flight unit and a dynamic model of a tracked travel unit.

[0106] Step S11: Establish a system model of a four-ducted flight unit. Define two coordinate systems, such as Figure 4 As shown, that is, the inertial coordinate system and body coordinate system According to the structural design characteristics, the structural form of the platform is simplified to a uniform and symmetrical structure, so that the origin of the body coordinate system coincides with the mass center of the amphibious vehicle.

[0107] Define the position p of the amphibious vehicle in the inertial coordinate system I =[x,y,z] T and posture η I =[φ,θ,ψ] T , where x is the x-axis coordinate of the amphibious vehicle in the inertial coordinate system; y is the y-axis coordinate of the amphibious vehicle in the inertial coordinate system; and z is the z-axis coordinate of the amphibious vehicle in the inertial coordinate system. Define the rotational angular velocity ω of the amphibious vehicle in the body coordinate system. B =[p,q,r] T and linear velocity v B =[u,v,w] T , where u is the linear velocity of the amphibious vehicle along the x-axis in the inertial coordinate system; v is the linear velocity of the amphibious vehicle along the y-axis in the inertial coordinate system; and w is the linear velocity of the amphibious vehicle along the z-axis in the inertial coordinate system. The system model of the four-ducted flight unit is expressed as follows.

[0108]

[0109] in, is the acceleration of the amphibious vehicle in the x-axis direction; is the acceleration of the amphibious vehicle in the y-axis direction; is the acceleration of the amphibious vehicle in the z-axis direction; g is the acceleration due to gravity; u t is the time input; u φ is the roll direction input; u θ is the yaw direction input; u ψ is the pitch direction input; m is the total mass of the amphibious vehicle; φ is the roll angle of the amphibious vehicle; θ is the yaw angle of the amphibious vehicle; ψ is the pitch angle of the amphibious vehicle; J=diag{J xx ,J yy ,J zz} is the moment of inertia of the amphibious vehicle relative to the inertial coordinate axis; is the angular acceleration of the amphibious vehicle around the x-axis; is the angular acceleration of the amphibious vehicle around the y-axis; is the angular acceleration of the amphibious vehicle around the z axis; r is the pitch angular rate of the amphibious vehicle; p is the roll angular rate of the amphibious vehicle; q is the yaw angular rate of the amphibious vehicle; J r is the moment of inertia of the motor; w represents the speed difference between the i-th motors with opposite rotation directions,

[0110] Step S12: Establish the dynamic model of the tracked driving unit. Since the ground driving unit of the amphibious vehicle is driven by the track differential, the interaction between the track and the ground is relatively complex. To simplify the calculation, the model is simplified while reflecting the necessary dynamic characteristics of the platform. Figure 5 As shown, is the inertial coordinate system of the amphibious vehicle, F l is the driving force of the left driving wheel of the amphibious vehicle, F r is the driving force of the right driving wheel of the amphibious vehicle, F i F is the slope resistance of the tracks on both sides of the amphibious vehicle, f F is the rolling resistance of the tracks on both sides of the amphibious vehicle, W is the air resistance on the surface of the amphibious vehicle, and v is the longitudinal movement speed of the amphibious vehicle.

[0111] Because the driving wheel torque is only controlled by the motor output torque. Therefore, the driving wheel driving force F on the left and right sides is l 、F r The method is as follows.

[0112]

[0113] Among them, T l is the output torque of the left motor of the amphibious vehicle, T r is the output torque of the right motor of the amphibious vehicle, and r is the rotation radius of the driving wheel of the amphibious vehicle.

[0114] Since the tracked platform has a small contact area with air resistance and a small effect on the total resistance, air resistance is ignored to simplify the subsequent modeling process. Based on the above equation, the dynamic model of the tracked travel unit is established, and its expression is as follows.

[0115]

[0116] Among them, T l is the output torque of the left motor of the amphibious vehicle; T r is the output torque of the right motor of the amphibious vehicle; intermediate variable β=sinα+fcosα, α is the slope of the ramp, f is the rolling resistance coefficient; r l is the driving wheel rotation radius of the amphibious vehicle; δ is the rotation mass conversion coefficient; The speed of the amphibious vehicle.

[0117] In a specific embodiment, step S2 specifically includes the following contents.

[0118] In the flight systems of amphibious vehicles, a ducted fan propulsion system uses a fan placed within an annular duct to generate thrust or lift. Its compact structure, low aerodynamic noise, and excellent operational safety are key features. Furthermore, at the same power consumption, it generates greater thrust than an open propeller of the same diameter, making it a highly efficient and safe propulsion system.

[0119] However, the aerodynamic characteristics of ducted fans are relatively complex. Due to the mutual influence between the ducted propellers, there is a vortex at the trailing edge of the moving blades, and this vortex will generate downwash velocity and downwash flow. In an unconstrained environment, the downwash flow of the blades is in a contraction state. Theoretically, the slipstream area at infinity is half the area of ​​the propeller disk. Figure 6 The three effects shown are: when the rotor outlet approaches the constrained environment (ground), the downwash flow is deflected radially outward, affecting the aerodynamic performance of the flight unit. This aerodynamic effect is called the ground effect; when the rotor inlet approaches the constrained environment (ceiling), the inflow of the rotor also has a significant radial deflection, affecting the aerodynamic performance of the flight unit, which is called the ceiling effect; when the side of the rotor approaches the constrained environment (vertical wall), the flow state of the rotor near the wall is different from that of the side away from the wall, which also affects the aerodynamic performance of the flight platform, which is called the wall effect.

[0120] Because the three effects mentioned above significantly influence the lift variations of amphibious vehicles, they also need to be considered as spatial dynamic constraints. Therefore, these spatial constraints must be considered when establishing the dynamic boundary constraints for the four-ducted flight unit. The aerodynamic characteristics of the ducted propellers must be analyzed, and a path planning algorithm for the amphibious vehicle under spatial constraints must be established. Therefore, the dynamic boundary constraints for the four-ducted flight unit are expressed as follows.

[0121] f p =λ dc f dc +λ dv f dv +λ da f da .

[0122] Among them, λ dc is the collision weight coefficient; λ dv is the weight coefficient of the speed dynamics feasibility constraint; da is the acceleration dynamics feasibility constraint weight coefficient; dc ,λ dv ,λ da , depends on the physical structural parameters of the ducted propeller, f dc is the collision cost between the ducted propeller structure and the obstacle, which increases rapidly if the trajectory approaches the obstacle; f dv is the rotation speed penalty of the ducted propeller, f da is the rotational acceleration cost of the ducted propeller. Considering the mutual interference between the propellers, the relationship between the additional aerodynamic force of the four modular ducts and the distance to the surrounding environment can be expressed as follows.

[0123]

[0124] Where σ = h / R, h is the distance between the amphibious vehicle and the surrounding wall environment, and R is the spiral radius. When σ > 7R, the influence of the three effects can be ignored.

[0125] In a specific embodiment, step S4 specifically includes: generating initial control points of the three-dimensional state quantity using the A* algorithm based on the original data, the dynamic model of the amphibious vehicle and the dynamic boundary constraints; and determining the shortest path for the amphibious vehicle to travel based on the initial control points.

[0126] Specifically, a hybrid A* method is used to introduce motion primitives to calculate the cost value in the land-air motion process. The search loop process is similar to the standard A* algorithm, where Open() and Close() represent open sets and closed sets. According to the starting coordinates and end coordinates of the land-air amphibious vehicle, the starting coordinates are added to the open set Open(). Compared with a straight line, the motion primitive of the land-air amphibious vehicle is generated through the land-air amphibious vehicle dynamics model and dynamic boundary constraints, and this motion primitive is used as a trajectory segment. The structure Node is used to record the motion primitive, the grid information of the primitive end position, and the process cost g goal and the total cost f goal .

[0127] Then, the node with the lowest cost in Open() is selected as the current node to generate the next primitive. This process is iteratively expanded across the grid map, ultimately generating the initial control point for the three-dimensional state at the final coordinate, thereby obtaining the shortest path for the amphibious vehicle. Simultaneously, the safety and dynamic traversability of the generated primitive curve are checked using Check() throughout the entire process. During the motion primitive generation process, the trajectory of the amphibious vehicle in one direction is first calculated based on the differential flatness of the amphibious vehicle. This trajectory is represented by a time-parameterized polynomial as follows.

[0128]

[0129] Among them, p x (t) refers to the state component in the x-axis direction at time t; the state vector p(t) = [p(t) T , p'(t) T ] T . Assume that the control input a(t) = p 2 (t)∈[-a max , a max ]. Therefore, the state space model can be defined as follows.

[0130] p x ′=Ap x +Ba.

[0131]

[0132] I3 is a standard unit three-dimensional matrix, and the general solution of the state equation is expressed as follows.

[0133]

[0134] Therefore, as long as the initial state p(0) and the control input a(t) are given, the trajectory of the amphibious vehicle on a plane can be calculated. Since the amphibious vehicle has two motion states, the path nodes in the path graph can be calculated based on the height pz It is divided into air nodes, ground nodes and take-off and landing nodes. When the motion primitive node is extended from the air, v is set for the ground node. z =0,a z = 0 to generate ground primitives, set v z = 0, az≠0 to generate takeoff and landing primitives. In order to find a trajectory with optimal time and control cost, the cost of the trajectory is defined as follows.

[0135]

[0136] Where J(a) is the heuristic cost function; T is the time it takes to move from the current position to the target point; λ state is the weight coefficient corresponding to different motion states. By setting a lower λ for the ground primitive state , the more ground nodes a trajectory contains, the smaller the corresponding trajectory cost will be, which is consistent with the fact that the cost of amphibious vehicles moving on the ground is relatively small. According to the concept of A* algorithm, using g n To represent the actual cost from the starting state to the current state. Therefore, the optimal path consists of J primitives, g n The calculation of is as follows.

[0137]

[0138] By applying the Pontryagin minimum principle, the optimal result J of the heuristic cost is generated. * (T) is expressed as:

[0139]

[0140] in, is the optimal kinematic trajectory; α u is the third-order coefficient; t is the time parameter; β μ is the coefficient of the second-order term; v μc is the velocity of the current point; p μc is the position of the current point; T is the movement time from the current position to the target point; p μg is the position of the target point; v μg is the velocity of the target point; J * (T) is the optimal heuristic cost.

[0141] J * (T) is the cost defined by formula J(a). In order to find the time T that minimizes the cost, J * α in (T) u , β μ Replace with v μc 、p μc 、v μg、p μg and t function, and finally through J * (T) Derivative, calculate the time T when the heuristic cost reaches the extreme value min This application uses J * (T min ) as a heuristic h c Finally, f c The definition is as follows.

[0142] f c =g c +h c =g c +J * (T min ).

[0143] In a specific embodiment, step S5 specifically includes: calculating the cost of the amphibious vehicle based on the shortest path; calculating the state augmentation cost by weighted summing the cost of the amphibious vehicle, and constructing a state augmentation objective function; based on the state augmentation objective function, converting the initial control point into a control point of a four-dimensional state quantity to obtain a converted control point; based on the converted control point, calculating the cost of the amphibious vehicle using the elastic band principle; calculating the total kinematic cost of the amphibious vehicle by weighted summing the cost of the amphibious vehicle; optimizing the shortest path of the amphibious vehicle based on the total kinematic cost of the amphibious vehicle to generate an optimized path for the amphibious vehicle.

[0144] In practice, when switching between land and air states, the switching process relies on information collected from sensors equipped on amphibious vehicles to determine the state of the environment. A set of instructions are then executed based on the information collected at the same time to complete the process safely. On the one hand, such a process consumes the limited computing resources carried by amphibious vehicles. On the other hand, the risk of encountering dangerous situations on the ground is generally higher. With this in mind, this application proposes to perform fewer mode switches during travel. Three performance indicators are considered in the design of the state augmentation: distance cost F d , energy consumption cost F e and mode switching cost F s , the calculation process of the state augmentation is as follows Figure 7 shown.

[0145] Distance cost F d : Distance is one of the main criteria considered in path planning. In most cases, the shorter the path, the better. The driving distance is defined as the sum of the Euclidean distances of the n+1 path polygons, where the length of the line segment is denoted by l. j (j=1, 2, ..., n+1), and the calculation formula is as follows.

[0146]

[0147] Among them, F d is the distance cost of amphibious vehicles; l j is the length of the jth line segment in the shortest path, where the line segment is the distance between two adjacent initial control points.

[0148] Energy consumption cost F e : The energy consumed per unit distance traveled on the ground and in the air is different. Generally speaking, due to high wind resistance and high power consumption, air travel is less efficient. Set the cost of amphibious vehicles on the ground ω land =1 and the cost of amphibious vehicles in flight ω air =5, meaning the energy consumption per unit of distance traveled on the ground and in the air is 1 and 5 units, respectively. Energy here doesn't necessarily refer to a physical meaning, but rather represents the relative cost of travel. Its calculation formula is as follows.

[0149] F e =ω land ∑l land +ω air ∑l air .

[0150] Among them, F e is the energy consumption cost of the amphibious vehicle; ω land is the cost of the amphibious vehicle when traveling on the ground; ∑l land is the sum of the lengths of the multi-line segments of the amphibious vehicle when it is traveling on the ground; ω air is the cost of the amphibious vehicle in flight; ∑l air It is the sum of the lengths of the polyline segments of the amphibious vehicle when it is flying.

[0151] In the above formula, ω land and ω air , is calculated using the following formula.

[0152]

[0153] Among them, F land is the traction force of the amphibious vehicle when it is traveling on the ground; v land F is the speed of the amphibious vehicle when traveling on the ground; air is the traction force of the amphibious vehicle when flying; v air It is the speed of an amphibious vehicle when flying.

[0154] Mode switching cost F sAs mentioned above, the cost of mode switching is an important factor to consider. To quantitatively evaluate the cost, the following formula is used to calculate it.

[0155] F s =N sup C sup +N sdown C sdown .

[0156] Among them, F s is the mode switching cost of the amphibious vehicle; N sup The number of times ground travel is converted into flight travel; C sup Cost of converting ground travel to flight travel; N sdown The number of ground trips converted from flight trips; C sdown The cost of converting ground travel to flight travel.

[0157] Finally, using the state augmentation cost F = ω1F d +ω2F e +ω3F s ,According to the flight experience of typical scenarios, a state quantity threshold FT can be set for the augmented state quantity, and the state augmentation quantity objective function can be obtained as follows.

[0158]

[0159] Among them, W is the state augmentation objective function; 1 is the amphibious vehicle in the flight state; 0 is the amphibious vehicle in the ground driving state; F is the state augmentation cost; F T The state augmentation threshold.

[0160] Specifically, during the trajectory optimization process, the initial control point of the three-dimensional state quantity is Q'={Q'0,Q'1,...,Q' N}, generate pb order B-spline curve (uniform B-spline curve is used in this paper), the initial control point Q' of each three-dimensional state quantity N =(x' i ,y' i ,z' i ). The state augmentation W consisting of 0 or 1 i Parameterized as a continuous curve between 0 and 1, the initial control point is transformed into the control point of the four-dimensional state quantity, and each control point after transformation is Q i =(x i ,y i ,z i ,W i ).

[0161] To optimize the trajectory, the following cost function is used: f1 = λ s fs +λ c f c +λ f (f v +f a ).

[0162] Where f1 is the total kinematic cost of the amphibious vehicle; s ,λ c ,λ f are weight values; f s is the smoothness cost of amphibious vehicles; f c is the collision cost of amphibious vehicles; f v is the speed feasibility cost of the amphibious vehicle; f a The acceleration feasibility cost of amphibious vehicles.

[0163] According to the elastic band principle, the smoothness cost and collision cost calculation formula are as follows.

[0164] The calculation formula of the smoothness cost is as follows.

[0165]

[0166] Among them, f s is the smoothness cost of the amphibious vehicle; N is the total number of initial control points; p b is the order of the B-spline curve; Q i+1 is the control point after the i+1th transformation; Q i is the control point after the i-th transformation; Q i-1 is the control point after the i+1th transformation.

[0167] The calculation formula of the collision cost is as follows.

[0168]

[0169] Among them, f c is the collision cost of amphibious vehicles; d(Q i ) is the control point Q after transformation i Distance to obstacles; d thr The distance between the amphibious vehicle and the obstacle.

[0170] When the speed or acceleration of an amphibious vehicle exceeds the maximum limit, the speed or acceleration in each direction is constrained using the following formula. The speed feasibility cost is calculated as follows.

[0171]

[0172] Among them, f vis the speed feasibility cost of the amphibious vehicle; μ is the {x, y, z} direction of the control point of the amphibious vehicle after the transformation; The control point Q of the amphibious vehicle after the transformation i Speed; v max The maximum speed of an amphibious vehicle.

[0173] The calculation formula of acceleration feasibility cost is as follows.

[0174]

[0175] Among them, f a The acceleration feasibility cost of amphibious vehicles; The control point Q of the amphibious vehicle after the transformation i acceleration; a max It is the maximum acceleration of an amphibious vehicle.

[0176] In a specific embodiment, step S6 specifically includes the following contents.

[0177] After the trajectory is optimized in step S5, the state augmentation objective function is binarized to convert it into a non-zero or 1 value representing the land and air state quantities. The current time is matched with the set point on the final generated trajectory, and then the transformed control point is used as the reference state information in the inertial coordinate system and sent to the controller of the land and air amphibious vehicle. All augmented control points include yaw angle, 3D position and state quantity. The land and air amphibious vehicle mode switching process is as follows: Figure 8 As shown, first, the optimal path is searched, and then the state augmentation objective function is binarized to determine the state augmentation W. current Is it 0? If so, the amphibious vehicle is in the ground driving state. If not, the amphibious vehicle is in the flying driving state. When the amphibious vehicle is in the ground driving state, determine the state augmentation amount W at the next moment. next Whether with W current If the ground driving state is equal, then the vehicle continues to move forward. If not, the amphibious vehicle switches to the ground mode. When the amphibious vehicle is in the flying state, the state augmentation amount W at the next moment is determined. next Whether with W current If they are equal, then determine whether the distance to the target point is greater than 1 meter. If not, switch the land-to-air mode. If the target point is greater than 1 meter, continue to fly. If the target point is greater than 1 meter, switch the land-to-air mode. Then, update the current state augmentation in real time to maintain the current state to the target point.

[0178] In a specific embodiment, the sensor and data communication architecture of the amphibious vehicle is as follows: Figure 9As shown in the figure, the camera in the sensor obtains feature points and depth point clouds, and the surrounding environment is mapped through the perception information to generate a grid map. The image obtained by the perception system and the inertial measurement unit (IMU) of the controller are used for navigation to generate the odometer. Planning is performed based on the grid map and the odometer, and local planning of the hybrid A* algorithm is performed respectively. The trajectory optimization system is used for optimization (Optimizer) to generate the state quantity (State) and B-spline trajectory (B-spline). The control instructions are then sent to the controller through the trajectory server to control the execution of the actuator (flight driving motor).

[0179] In a specific embodiment, three typical scenarios were designed to conduct actual test experiments on the effectiveness of the state-augmented amphibious vehicle path planning and mode switching method of the present application. By collecting experimental data and comparative analysis of vehicle movement in different scenarios, the effectiveness of the algorithm proposed in the present application was verified. The test environment is as follows: Figure 10 As shown. Among them, Figure 10 (a) is to verify the trafficability of the amphibious vehicle in the ground driving state, and the target point position of the autonomous driving task is (0, 0, 6.5m); Figure 10 (b) is to verify the high maneuverability of the amphibious vehicle in flight. The target point position of the autonomous flight mission is (0, 0, 8.5m); Figure 10 (c) is to verify the rapid passing performance of the amphibious vehicle in a narrow space using the path planning method proposed in this application in a typical indoor environment. The mission target point position of the autonomous amphibious planning is (0, 0, 8.5m). The amphibious vehicle first quickly reaches the obstacle that needs to be passed through through the ground driving mechanism, and then switches the land-air mode, passes through the obstacle through autonomous planning and obstacle avoidance in the air, and finally switches back to the ground mode, reaching the designated location under the optimal conditions of energy consumption and travel time cost.

[0180] According to the three test environments designed above, a simulation test was first conducted. After using the above method, the simulation effect of the amphibious vehicle's driving trajectory is as follows: Figure 11 As shown, Figure 11 (a) shows the simulation result of the optimal path in the ground obstacle crossing environment; Figure 11 (b) is the simulation result of the optimal path in the aerial obstacle crossing environment; Figure 11 (c) in the figure shows the simulation result of the optimal path in a three-dimensional narrow space environment.

[0181] Then, the proposed solution of this application was tested experimentally. Figure 12In (a), since the obstacle height is within the passable range of the ground driving system, there is no need to switch states. Therefore, the state planning quantities in this scenario are all driving mode state quantities. In the path length comparison results generated by the algorithm proposed in this application and the original algorithm, when the target path length is set to 6.5m, the average driving path length of the original algorithm is 7.859m, and the average driving path length of the algorithm proposed in this application is 7.247m. At the same time, the displacement deviation of the proposed algorithm in the Y and Z directions is smaller, the path planning trajectory is smoother, the driving distance is shorter, and the planned trajectory is better. Figure 12 In (b), during the planning process of the amphibious vehicle, there is an obstacle in the middle of the road that is impassable for ground driving, and a land-air switch is required. Under the condition of setting the target point distance to 8.5m, the average path planning length of the original method is 17.052m, and the average path length generated by the proposed algorithm is 15.672m, which is nearly 2 meters shorter than the original algorithm. Similarly, the proposed algorithm has smaller displacement deviations in the Y and Z directions, the path planning trajectory is smoother, the driving distance is shorter, and the planned trajectory is better. Figure 12 In figure (c), an amphibious vehicle traverses a narrow square space. In the original algorithm, the trajectory of the amphibious vehicle after passing through a window fails to return to the ground in time, resulting in a poorly planned trajectory. Compared to the original algorithm, the proposed algorithm advances the transition between land and air. Furthermore, by controlling the trajectory and landing position of the platform after passing the obstacle based on its state information, the proposed algorithm makes the overall trajectory smoother and reduces the planned trajectory length.

[0182] Figure 13 It shows the real-time movement speed changes of amphibious vehicles in different scenarios. Figure 13 In (a), since the amphibious vehicle has state information, when the state is known to be in the driving state, there is no need to consider state switching. The road conditions are within the traffic capacity of the amphibious vehicle. Therefore, the maximum speed can be continuously increased. The average maximum speed reaches 0.590m / s, which is 0.055m / s higher than the original 0.535m / s. The amphibious vehicle quickly overcomes the obstacle and reaches the end. Figure 13 In (b), the amphibious vehicle needs to bypass an impassable obstacle in the middle of the road, and it needs to switch states. However, the amphibious vehicle can know that it needs to switch states in the middle process through the state quantity. In the process where no state switching occurs, the amphibious vehicle can quickly increase the driving speed and then quickly reduce the speed when it needs to switch states in order to enter the mode switching state. Figure 13In (c), the amphibious vehicle, leveraging its high maneuverability and ability to navigate narrow spaces, achieves faster speeds than the original algorithm, ultimately reaching the target point. Comparing the average speeds across the three scenarios shows that the proposed method achieves a higher maximum speed, improves the average speed throughout the entire process, and significantly shortens the travel time.

[0183] Table 1 Average moving distance (m) under different environments

[0184] environment end Original method This application method Improved results 1 (6.5,0,0) 7.859 7.247 7.79％ 2 (8.5,0,0) 17.052 15.672 8.09％ 3 (8.5,0,0) 18.539 16.172 12.77％

[0185] Table 2 Average speed under different environments (m / s)

[0186] environment Original method This application method Improved results 1 0.535 0.590 10.28％ 2 0.714 0.797 11.62％ 3 0.669 0.755 12.86％

[0187] Table 3 Exercise time (s) in different environments

[0188] environment Original method This application method Improved results 1 11.750 10.411 11.40％ 2 47.717 39.066 18.13 3 48.466 37.104 23.44％

[0189] Tables 1-3 compare the average distance traveled, average speed, and travel time in different environments with and without state augmentation. Environment 1 corresponds to a ground obstacle crossing environment; Environment 2 corresponds to an aerial obstacle crossing environment; and Environment 3 corresponds to a three-dimensional confined space environment. Data analysis revealed that in Environment 1, the amphibious vehicle's ground driving system features a highly passable tracked structure. Combined with the path planning algorithm with state augmentation, the amphibious vehicle can quickly navigate obstacles less than 9 cm in driving mode. Comparing experimental data with and without state augmentation, it was found that path planning with state augmentation reduced travel distance by 7.79%, increased average speed by 10.28%, and shortened travel time by 11.40%. In Environment 2, because there is an impassable obstacle in front of the amphibious vehicle, it needs to switch between land and air modes during the journey. The path planning algorithm with state augmentation includes platform state information in the instructions, so the platform can quickly reach the land-air switching position and quickly take off to cross the obstacle. According to the data, the platform's movement distance during this process was reduced by 8.09%, the average driving and flying speeds increased by 11.62%, and the movement time was shortened by 18.13%. In Environment 3, when the platform passes through a narrow and confined space, it needs to be affected by various aerodynamic effects generated by surrounding obstacles. At the same time, the amphibious vehicle needs to switch between land and air modes to pass. The amphibious vehicle adopts the method of this application, and the distance traveled is reduced by 12.77%, the speed is increased by 12.86%, and the passage time is shortened by 23.44%.

[0190] The beneficial effects of the state-augmented amphibious vehicle path planning and mode switching method proposed in this application are mainly manifested in: this application proposes an amphibious path planning that takes into account the state augmentation amount, simplifies the state switching strategy, shortens the state switching time, reduces energy consumption, and improves driving efficiency. At the same time, the amphibious unmanned vehicle not only has coordinates but also the driving mode of the amphibious vehicle during the path driving process, and can predict the driving state of the amphibious vehicle at the next moment, so that the amphibious vehicle can enter the driving state transition in advance, thereby improving the safety of the amphibious vehicle during driving.

[0191] In an exemplary embodiment, a state-augmented amphibious vehicle path planning and mode switching system is provided, comprising the following modules.

[0192] The dynamic model construction module is used to construct the dynamic model of the land and air amphibious vehicle; the land and air amphibious vehicle includes a four-ducted flight unit and a tracked driving unit.

[0193] The dynamic boundary constraint condition construction module is used to construct the dynamic boundary constraint conditions of the four-ducted flight unit.

[0194] The acquisition module is used to obtain the original data of the amphibious vehicle's driving path; the original data includes: starting coordinates and end point coordinates.

[0195] The shortest path generation module is used to generate the shortest path for amphibious vehicles based on the original data, the dynamic model of the amphibious vehicle and the dynamic boundary constraints, using the A* algorithm.

[0196] The optimized path generation module is used to optimize the shortest path based on the state augmentation quantity to obtain the optimized path for the amphibious vehicle and complete the path planning of the amphibious vehicle; the state augmentation quantity represents the driving mode of the amphibious vehicle.

[0197] The driving mode switching module is used to switch the driving mode of the amphibious vehicle based on the optimized path.

[0198] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0199] This document uses specific examples to illustrate the principles and implementation methods of this application. The description of the above examples is only intended to help understand the method and core concept of this application. At the same time, for those skilled in the art, based on the concept of this application, there may be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A state-augmented amphibious vehicle path planning and mode switching method, characterized in that: The state-augmented amphibious vehicle path planning and mode switching method includes: Constructing a dynamic model of an amphibious vehicle comprising a four-ducted flight unit and a tracked travel unit; Construct the dynamic boundary constraints of the four-ducted flight unit; Obtaining original data of the amphibious vehicle's travel path; the original data includes: starting coordinates and end coordinates; Based on the original data, the dynamic model of the amphibious vehicle and the dynamic boundary constraints, an A* algorithm is used to generate the shortest path for the amphibious vehicle to travel; The shortest path is optimized based on a state augmentation quantity to obtain an optimized path for the amphibious vehicle, thereby completing the path planning of the amphibious vehicle; the state augmentation quantity represents the driving mode of the amphibious vehicle; The driving mode of the amphibious vehicle is switched based on the optimized path.

2. The state-augmented amphibious vehicle path planning and mode switching method according to claim 1, characterized in that: The dynamic model of the amphibious vehicle includes a system model of a four-ducted flight unit and a dynamic model of a tracked travel unit; The expression of the system model of the four-ducted flight unit is: in, is the acceleration of the amphibious vehicle in the x-axis direction; is the acceleration of the amphibious vehicle in the y-axis direction; is the acceleration of the amphibious vehicle in the z-axis direction; g is the acceleration due to gravity; u t is the time input; u φ is the roll direction input; u θ is the yaw direction input; u ψ is the pitch direction input; m is the total mass of the amphibious vehicle; φ is the roll angle of the amphibious vehicle; θ is the yaw angle of the amphibious vehicle; ψ is the pitch angle of the amphibious vehicle; J=diag{J xx ,J yy ,J zz } is the moment of inertia of the amphibious vehicle relative to the inertial coordinate axis; is the angular acceleration of the amphibious vehicle around the x-axis; is the angular acceleration of the amphibious vehicle around the y-axis; is the angular acceleration of the amphibious vehicle around the z axis; r is the pitch angular rate of the amphibious vehicle; p is the roll angular rate of the amphibious vehicle; q is the yaw angular rate of the amphibious vehicle; J r is the moment of inertia of the motor; w represents the speed difference between the i-th motors with opposite rotation directions; The dynamic model of the crawler travel unit is expressed as follows: Among them, T l is the output torque of the left motor of the amphibious vehicle; T r is the output torque of the right motor of the amphibious vehicle; intermediate variable β=sinα+fcosα, α is the slope of the ramp, f is the rolling resistance coefficient; r l is the driving wheel rotation radius of the amphibious vehicle; δ is the rotation mass conversion coefficient; The speed of the amphibious vehicle.

3. The state-augmented amphibious vehicle path planning and mode switching method according to claim 1, characterized in that: The dynamic boundary constraints of the four-ducted flight unit are expressed as: f p =λ dc f dc +λ dv f dv +λ da f da ; Among them, λ dc is the collision weight coefficient; λ dv is the weight coefficient of the speed dynamics feasibility constraint; da is the acceleration dynamics feasibility constraint weight coefficient; f dc is the collision cost between the ducted propeller structure and the obstacle; f dv is the rotation speed penalty of the ducted propeller, f da is the rotational acceleration cost of the ducted propeller.

4. The state-augmented amphibious vehicle path planning and mode switching method according to claim 1, characterized in that: Based on the original data, the dynamic model of the amphibious vehicle, and the dynamic boundary constraints, an A* algorithm is used to generate the shortest path for the amphibious vehicle to travel, specifically including: Based on the original data, the dynamic model of the amphibious vehicle and the dynamic boundary constraints, an A* algorithm is used to generate initial control points of a three-dimensional state quantity; The shortest path for the amphibious vehicle to travel is determined based on the initial control point.

5. The state-augmented amphibious vehicle path planning and mode switching method according to claim 4, characterized in that: The shortest path is optimized based on the state augmentation amount to obtain the optimized path for the amphibious vehicle, specifically including: Calculating the cost of the amphibious vehicle based on the shortest path; Calculating the state augmentation cost by weighted summing of the land-air amphibious vehicle costs and constructing a state augmentation objective function; Based on the state augmentation objective function, the initial control point is transformed into the control point of the four-dimensional state quantity to obtain the transformed control point; Based on the transformed control points, the cost of amphibious vehicles is calculated using the elastic band principle; Calculating a total kinematic cost of the amphibious vehicle by taking a weighted sum of the costs of the amphibious vehicle; Based on the total kinematic cost of the amphibious vehicle, the shortest path of the amphibious vehicle is optimized to generate the optimized path of the amphibious vehicle.

6. The state-augmented amphibious vehicle path planning and mode switching method according to claim 5, characterized in that: The cost of the land-air amphibious vehicle includes: distance cost, energy consumption cost and mode switching cost; The distance cost is calculated as: Among them, F d is the distance cost of amphibious vehicles; l j is the length of the jth line segment in the shortest path, where the line segment is the distance between two adjacent initial control points; The formula for calculating energy consumption cost is: F e =ω land v.l. land +oh air ∑l air ; Among them, F e is the energy consumption cost of the amphibious vehicle; ω land is the cost of the amphibious vehicle when traveling on the ground; ∑l land is the sum of the lengths of the multi-line segments of the amphibious vehicle when it is traveling on the ground; ω air is the cost of the amphibious vehicle in flight; ∑l air is the sum of the lengths of the multi-line segments of the amphibious vehicle when it is flying; The calculation formula for mode switching cost is: F s =N sup C sup +N sdown C sdown ; Among them, F s is the mode switching cost of the amphibious vehicle; N sup The number of times ground travel is converted into flight travel; C sup Cost of converting ground travel to flight travel; N sdown The number of ground trips converted from flight trips; C sdown The cost of converting ground travel to flight travel.

7. The state-augmented amphibious vehicle path planning and mode switching method according to claim 5, characterized in that: The calculation formula for the state augmentation cost is: F=ω1F d +ω2F e +ω3F s ; The expression of the state augmentation objective function is: Among them, W is the state augmentation objective function; 1 is the amphibious vehicle in the flight state; 0 is the amphibious vehicle in the ground state; F is the state augmentation cost; F T The state augmentation threshold.

8. The state-augmented amphibious vehicle path planning and mode switching method according to claim 5, characterized in that: The cost of the amphibious vehicle includes: smoothness cost, collision cost, speed feasibility cost and acceleration feasibility cost; The calculation formula of smoothness cost is: Among them, f s is the smoothness cost of the amphibious vehicle; N is the total number of initial control points; p b is the order of the B-spline curve; Q i+1 is the control point after the i+1th transformation; Q i is the control point after the i-th transformation; Q i-1 is the control point after the i+1th transformation; The calculation formula for the collision cost is: Among them, f c is the collision cost of amphibious vehicles; d(Q i ) is the control point Q after transformation i Distance to obstacles; d thr is the distance between the amphibious vehicle and the obstacle; The calculation formula of speed feasibility cost is: Among them, f v is the speed feasibility cost of the amphibious vehicle; μ is the {x, y, z} direction of the control point of the amphibious vehicle after the transformation; The control point Q of the amphibious vehicle after the transformation i Speed; v max The maximum speed of an amphibious vehicle; The calculation formula of acceleration feasibility cost is: Among them, f a The acceleration feasibility cost of amphibious vehicles; The control point Q of the amphibious vehicle after the transformation i acceleration; a max It is the maximum acceleration of an amphibious vehicle.

9. The state-augmented amphibious vehicle path planning and mode switching method according to claim 5, characterized in that: The total kinematic cost of an amphibious vehicle is calculated as: f1=λ s f s +λ c f c +λ f (f v +f a ); Where f1 is the total kinematic cost of the amphibious vehicle; s ,λ c ,λ f are weight values; f s is the smoothness cost of amphibious vehicles; f c is the collision cost of amphibious vehicles; f v is the speed feasibility cost of the amphibious vehicle; f a The acceleration feasibility cost of amphibious vehicles.

10. A state-augmented amphibious vehicle path planning and mode switching system, characterized in that: The state-augmented amphibious vehicle path planning and mode switching method according to any one of claims 1 to 9, wherein the state-augmented amphibious vehicle path planning and mode switching system comprises: A dynamic model building module, used to build a dynamic model of an amphibious vehicle; the amphibious vehicle includes a four-ducted flight unit and a tracked travel unit; Dynamic boundary constraint condition construction module, used to construct the dynamic boundary constraint conditions of the four-ducted flight unit; An acquisition module is used to acquire the original data of the amphibious vehicle's travel path; the original data includes: starting coordinates and end point coordinates; a shortest path generation module, configured to generate a shortest path for the amphibious vehicle using an A* algorithm based on the original data, the dynamic model of the amphibious vehicle, and the dynamic boundary constraints; an optimized path generation module, configured to optimize the shortest path based on a state augmentation quantity to obtain an optimized path for the amphibious vehicle, thereby completing path planning for the amphibious vehicle; the state augmentation quantity represents the driving mode of the amphibious vehicle; A driving mode switching module is used to switch the driving mode of the amphibious vehicle based on the optimized path.

Images