A method and system for adaptive adjustment of a driving carrier platform under cooperation of a UAV

Abstract

The application relates to the technical field of intelligent cooperative control, and discloses a driving and carrying platform self-adaptive adjustment method and system under the cooperation of unmanned aerial vehicles. The method comprises the following steps: constructing an unmanned aerial vehicle flight path set; constructing a field of view overlap starting area set F; adopting a double-source synchronous locking and weight progressive fusion mechanism to output a unified perception state set; introducing a dynamic gain adjustment factor to generate a control instruction sequence; executing compressed instruction mutation fluctuation; and applying a smooth control instruction set. Compared with the single perception link based on the prior art, especially in a large scene such as a long-distance transportation road or a large container terminal yard, a running scene in which a section of path needs to be covered by multiple unmanned aerial vehicles in relay, the technical problem that the adaptive control of the unmanned aerial vehicle cooperative driving platform is difficult to realize. Since the application constructs a dynamic gain adjustment and fuses an instruction compression strategy of a double-sided limiting mechanism, the driving track adjustment continuity and stability in the cooperative perception scene are improved.

Description

An adaptive adjustment method and system for a driving and transport platform in collaboration with unmanned aerial vehicles (UAVs) Technical Field

[0001] This invention relates to the field of intelligent collaborative control technology, and in particular to an adaptive adjustment method and system for a driving and transport platform under unmanned aerial vehicle (UAV) collaboration. Background Technology

[0002] Currently, in complex transportation scenarios such as mining area transport roads, container terminal yards, large ports, and construction sites, dynamic relay observation systems based on multiple drones are increasingly being used for continuous monitoring and path guidance of driving transport platforms. However, in such systems, due to the wide monitoring area, limited overlap of sensing range, and the limited endurance and bandwidth of drones, sensing tasks often require multiple drones to work together in relay, leading to problems with sensing synchronization and control continuity during the relay switching process.

[0003] For example, in existing technologies, when the lead drone is about to leave the perception zone after completing its mission, while the relay drone has not yet completed a stable perception handover with the target platform, a "gap" or "abrupt change" in control commands can easily occur. This manifests as sudden changes in steering angle and speed fluctuations, severely impacting the operational stability and safety of the carrier platform. Existing methods often rely on fixed, preset handover times or signal switching mechanisms based on global synchronization, lacking a comprehensive assessment of dynamic factors such as mission redundancy between drones, airspace clearance time, remaining battery power, and bandwidth status. This makes it difficult to achieve precise scheduling and smooth connection of control commands during the perception handover process. Existing technologies cannot fully meet the comprehensive requirements for continuity of control information, real-time handover scheduling, and smooth platform adjustment in situations where multiple drones collaborate in perception handover across regions and lanes.

[0004] Therefore, there is an urgent need for an adaptive adjustment method for a drone-assisted driving platform that can achieve seamless transition and steady-state execution of platform control commands even under conditions of high uncertainty in perception relay switching, so as to improve platform operation stability and perception relay efficiency. Summary of the Invention

[0005] To address the aforementioned technical shortcomings, the present invention aims to propose an adaptive adjustment method for a drone-assisted driving platform. This method addresses the technical problem of difficulty in achieving adaptive control of a drone-assisted driving platform in existing technologies based on a single perception link, especially in large-scale scenarios such as long-distance transport roads or large container terminal yards where multiple drones are needed to relay coverage of a single path.

[0006] To solve the above-mentioned technical problems, the present invention adopts the following technical solution: The present invention provides an adaptive adjustment method for a driving and transport platform under UAV collaboration.

[0007] The adaptive adjustment method for the driving and transportation platform under UAV collaboration includes:

[0008] Step S10: Obtain target transportation route information and site topology information, and determine based on target transportation route information and site topology information. The flight path set of the collaborative UAV is used to perform the field of view overlap determination task based on the relay redundancy mapping mechanism based on the residual task window constraint, and output the field of view overlap starting region set F;

[0009] Step S20: Based on the set of overlapping starting regions F, a feature handover task is performed using a dual-source synchronous locking and weighted progressive fusion mechanism to output a unified perception state set. ;

[0010] Step S30: Based on the unified sensing state set A gain-adaptive buffering mechanism is used to generate a smooth control instruction set;

[0011] Step S40: Based on the smooth control instruction set, a limiting and suppression mechanism based on rate constraints is used to perform key control quantity limiting tasks, and the limiting and smooth control instruction set is output.

[0012] Step S50: Apply the amplitude-limiting smooth control instruction set to perform adaptive adjustment of the driving vehicle platform.

[0013] Preferably, in step S10, target transportation route information and site topology information are obtained, and determination is made based on the target transportation route information and site topology information. The steps involved in determining the field of view overlap and outputting the set of initial overlapping regions F from the flight paths of the collaborative UAVs include:

[0014] Step S101: Obtain target transportation route information and site topology information, and determine based on target transportation route information and site topology information. The set of flight paths of a collaborative UAV; obtain the first flight path in the set of flight paths. Remaining battery information for the drone Unit power consumption information Airspace occupancy permit time information Information on the estimated completion time of the current task Based on remaining battery power information Unit power consumption information Airspace occupancy permit time information Information on the estimated completion time of the current task The first method is constructed using a joint decision-making approach based on energy consumption feasible region constraints and space occupancy restrictions. Remaining mission window for drones Remaining task window This is used to characterize the redundant time range within which a UAV can continue to be used for relay observation after completing its main task.

[0015] Step S102: Based on the remaining task window Constructing candidate relay periods , In order to be with the first The start time of the candidate relay period for the j-th drone adjacent to the first drone; In order to be with the first The end time of the candidate relay period for the j-th drone adjacent to the first drone; , ;

[0016] Step S103: Jointly Flight altitude of the drone With depression angle Determine the ground field of view coverage radius United Nations Flight altitude of the drone With depression angle Determine the ground field of view coverage radius ; Get the The drone and the first Two drones If satisfied Then determine the candidate relay period. For the effective relay period, the effective relay period is regarded as the starting region of field of view overlap, and the final output is the set of starting regions of field of view overlap F.

[0017] Preferably, in step S20, based on the set of overlapping starting regions F, a feature handover task is performed using a dual-source synchronous locking and weighted progressive fusion mechanism to output a unified perception state set. The steps specifically include:

[0018] Step S201: Based on the set of overlapping field-of-view starting regions F, obtain the broadcast vehicle status set sent by the current leading UAV b to the relay UAV m. , , For location information, For speed information, For heading angle information;

[0019] Step S202: The relay drone m relays the vehicle status set according to the broadcast. Perform vehicle recognition and target locking tasks, and output local feedback perception results. ;

[0020] Step S203: Perceive results based on local feedback and broadcast vehicle status set A linear interpolation fusion method based on time-progressive smooth collaboration is used to obtain a unified perception state result. .

[0021] Preferably, in step S20, the unified perception state result is obtained. The formula is expressed as:

[0022]

[0023] in, Let be the fusion weight function at any time t. , This is the moment of relay switching; This is the effective relay period.

[0024] Preferably, in step S30, based on the unified sensing state set The steps for generating a smooth control instruction set using a gain adaptive buffering mechanism specifically include:

[0025] Based on unified sensing state set Generate initial control commands at any time t Simultaneously acquire historical control commands at time t-1. Based on initial control commands and historical control commands The current smoothing control command is constructed using a first-order IIR filter structure with weighted dynamic suppression. ;

[0026] Current smooth control command The formula is expressed as:

[0027]

[0028] in, The introduced dynamic gain suppression factor, adjusted according to the drone relay status, is defined as follows:

[0029]

[0030] in, Basic gain suppression factor, ; This is the gain adjustment intensity coefficient. .

[0031] Preferably, step S40, which involves executing the key control quantity limiting task based on a rate-constraint-based limiting and suppression mechanism using a smooth control instruction set, and outputting the limiting and smooth control instruction set, specifically includes:

[0032] Step S401: Obtain the current steering angle control command and the current acceleration control command from the smooth control command set, and obtain the preset desired steering angle control command and desired acceleration control command; construct the first control variable rate factor based on the current steering angle control command and the desired steering angle control command using residual analysis and normalization calculation methods. Based on the current acceleration control command and the desired acceleration control command, the second control variable rate factor is constructed using residual analysis and normalization calculation methods. ;

[0033] Step S402: Based on the first control variable rate factor Second control quantity rate factor The steering angle variation limiting threshold is constructed using the linear annealing mapping method. and acceleration rate limiting threshold ;

[0034] Step S403: Limiting threshold based on steering angle variation and acceleration rate limiting threshold A bilateral symmetric limiting function is used to compress the smoothing control instruction set, and the final output is a limiting smoothing control instruction set.

[0035] Preferably, in step S40, the formula for the bilateral symmetric limiting function is expressed as follows:

[0036]

[0037] in, Let be a bilaterally symmetric amplitude-limiting function at any time t; For smooth control of any instruction signal in the instruction set The original value at any time t; For smooth control of any instruction signal in the instruction set The control value at the previous time t-1; For command signals The rate limiting threshold is constructed at any time t, including the steering angle rate limiting threshold. and acceleration rate limiting threshold ; It is a symbolic function.

[0038] This invention also provides an adaptive adjustment system for a drone-assisted driving and transportation platform, comprising:

[0039] The route planning module is used to acquire target transportation route information and site topology information, and to determine the route based on the target transportation route information and site topology information. The flight path set of the collaborative UAV is used to perform the field of view overlap determination task based on the relay redundancy mapping mechanism based on the residual task window constraint, and output the field of view overlap starting region set F;

[0040] The state handover module is used to perform feature handover tasks based on the set of overlapping field-of-view regions F, employing a dual-source synchronous locking and weighted progressive fusion mechanism, and outputs a unified perception state set. ;

[0041] Instruction smoothing module, used for unified sensing state set A gain-adaptive buffering mechanism is used to generate a smooth control instruction set;

[0042] The amplitude limiting and suppression module is used to perform key control quantity limiting tasks based on the amplitude limiting and suppression mechanism based on the rate constraint of the smooth control instruction set, and outputs the amplitude limiting and smooth control instruction set;

[0043] The platform execution module is used to apply the amplitude-limiting smooth control instruction set to perform adaptive adjustment of the driving vehicle platform.

[0044] The present invention also provides an adaptive adjustment device for a drone-cooperative driving vehicle platform, comprising: a memory, a processor, and an adaptive adjustment program for a drone-cooperative driving vehicle platform stored in the memory and executable on the processor. When the drone-cooperative driving vehicle platform adaptive adjustment program is executed by the processor, it implements an adaptive adjustment method for a drone-cooperative driving vehicle platform.

[0045] The present invention also provides a computer program product, including an adaptive adjustment program for a driving vehicle platform under UAV collaboration, wherein the adaptive adjustment program for a driving vehicle platform under UAV collaboration implements the adaptive adjustment method for a driving vehicle platform under UAV collaboration when executed by a processor.

[0046] The beneficial effects of this invention are as follows: By constructing a relay redundancy mapping mechanism based on residual task window constraints and a dual-source synchronous locking fusion strategy, this invention can stably output a unified perception state under the condition of inconsistent field-of-view intersection areas, ensuring that control commands have complete and continuous data support during the dynamic relay of multi-machine perception tasks, and avoiding the problems of control command gaps, abrupt changes and jitter caused by the switching of a single perception source in the prior art.

[0047] This invention introduces a dynamic gain buffering mechanism and a rate limiting suppression mechanism. By dynamically suppressing and bilaterally compressing the control command sequence, it effectively alleviates the risk of control signal jumps caused by bandwidth bottlenecks and latency jitter during UAV frame switching. Compared with the traditional "frame-by-frame response" control method, it can significantly improve the continuity and feasibility of control command output while ensuring real-time response. Attached Figure Description

[0048] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0049] Figure 1 is a flowchart illustrating the first embodiment of an adaptive adjustment method for a driving and transport platform under UAV collaboration according to the present invention.

[0050] Figure 2 is a schematic diagram of dual-source synchronous locking and feature handover of the first embodiment of the adaptive adjustment method for a driving and transport platform under UAV cooperation of the present invention.

[0051] Figure 3 is a schematic diagram of the device for an adaptive adjustment method of a driving and transport platform under UAV collaboration according to the present invention. Detailed Implementation

[0052] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0053] Example 1: As shown in Figure 1, this is a flowchart illustrating the first embodiment of the adaptive adjustment method for the driving platform under UAV collaboration of the present invention. The first embodiment of the adaptive adjustment method for the driving platform under UAV collaboration of the present invention is presented.

[0054] In the first embodiment, the adaptive adjustment method for the driving and transport platform under UAV collaboration includes:

[0055] Step S10: Obtain target transportation route information and site topology information, and determine based on target transportation route information and site topology information. The flight path set of the collaborative UAV is used to perform the field of view overlap determination task based on the relay redundancy mapping mechanism based on the residual task window constraint, and output the field of view overlap starting region set F;

[0056] It should be noted that the "relay redundancy mapping mechanism based on residual task window constraints" refers to the schedulable capability that each collaborative UAV still possesses after completing its predetermined main task, which is quantitatively described by constructing a comprehensive residual capability index window. This residual task window not only includes basic parameters such as the UAV's remaining battery power, the estimated average power consumption per unit trajectory, the task return time constraint, and the airspace permission time boundary, but also further incorporates dynamic information such as the expected end time of the UAV's current task phase, the expected sensing load intensity, and the spatial distribution of the relay target area.

[0057] Understandably, the relay redundancy mapping mechanism built on the remaining task window allows task handover between UAVs to no longer rely on fixed preset spatial intersection points or static time windows, but can dynamically adjust the relay start interval according to the remaining capacity and relay requirements of different UAVs. By selecting relay combinations with higher spatiotemporal coupling within the redundancy window, this step can effectively reduce the interruption of the perception link caused by UAVs suddenly ending their tasks, returning to base, or short-term field of view drift, thereby improving the stability of subsequent feature handover and perception fusion.

[0058] It should be understood that, compared to the "fixed handover point triggering strategy" or "single time overlap judgment strategy" commonly used in traditional technologies, the innovation of this step lies in: by constructing the remaining task window, multiple real engineering conditions, such as the UAV's schedulable capabilities (e.g., battery level, power consumption, return-to-home constraints), task stage attributes (e.g., end time, flight path congestion), and airspace resource restrictions (e.g., permitted time periods, flight path conflicts), are simultaneously incorporated into the relay judgment. This multi-dimensional dynamic constraint method avoids the vulnerability of traditional methods where a deviation in a single parameter (e.g., battery prediction error or communication delay) can lead to overall handover failure, upgrading the relay judgment from "single-factor driven" to "multi-factor collaborative driven."

[0059] For example, in a test of a transportation route in a mining area, when the lead UAV was tracking a vehicle on a long slope, a sudden increase in wind speed caused a significant trajectory deviation, resulting in its expected completion time being tens of seconds earlier than predicted by the original model. If calculated according to the traditional preset relay point, the relay UAV would not be able to enter the field of view coverage in time, causing the vehicle target to be briefly unlocked. After adopting the residual task window constraint mechanism of this invention, the remaining battery power of the lead UAV, the degree of trajectory fluctuation, and the airspace clearance time are incorporated into the window model in real time, and its redundant window is dynamically compressed. At the same time, the relay UAV is automatically scheduled to an earlier handover area according to its own residual window. Finally, the set F of the field of view overlap starting areas is triggered in advance, and there is no obvious target loss phenomenon during the actual switching process, and the continuity of vehicle control commands remains stable. In subsequent rounds of comparative tests, the average proportion of short-term perception interruption in the scheme using traditional relay determination is significantly higher than that in the scheme of this invention. This fully demonstrates that the redundancy mapping and window constraint mechanism introduced in this step can effectively reduce risks and improve the relay success rate of collaborative UAVs under engineering conditions.

[0060] Step S20: Based on the set of overlapping starting regions F, a feature handover task is performed using a dual-source synchronous locking and weighted progressive fusion mechanism to output a unified perception state set. ;

[0061] It should be noted that the "dual-source synchronous locking and weighted progressive fusion mechanism" refers to the synchronous feature acquisition and handover fusion of the target area after the set F of the overlapping field of view is determined, relying on the dual sensing source inputs of the leading UAV and the relay UAV. In this mechanism, "dual-source synchronous locking" aims to ensure the consistency and reconstruction integrity of the sensing features (such as moving target trajectory, color distribution, spatial contour, etc.) on the sampling time axis by two UAVs synchronously observing the same target area within the spatiotemporal overlap window; while "weighted progressive fusion" constructs a dynamic weight coefficient vector based on the sharpness score, overlapping frame number, complementary view coverage, and historical reliability score of the two-source observation data, and performs progressive data fusion processing accordingly, finally outputting a unified sensing state set.

[0062] Understandably, the synchronous locking and weighted fusion mechanism of dual-source perception not only improves the continuity and accuracy of targets during feature handover but also avoids information gaps caused by the sudden loss of contact or target lock-off by the lead UAV. The weighted progressive strategy ensures that the lead UAV's perception weight automatically decays as it approaches return home or moves out of the field of view, while the relay UAV's perception weight gradually increases, thus giving the fusion process a "smooth transition" characteristic and preventing abrupt state jumps. This strategy significantly improves the temporal consistency and spatial integrity of the perception state set, laying a reliable perception foundation for subsequent target tracking, task command generation, and collaborative path control.

[0063] For example, as shown in Figure 2, the two ellipses represent the fields of view of the dominant and relay drones; the orange area in the middle is the starting region F where the fields of view overlap; the red dot represents the target object; and the black arrow indicates the direction of feature intersection. The dominant drone and the relay drone achieve synchronous locking within region F. Based on the trajectory features and edge contour information collected by the two drones, a weighted progressive fusion process is performed to ultimately form a continuous and consistent unified perception state set, avoiding the target drift problem during visual switching.

[0064] Step S30: Based on the unified sensing state set A gain-adaptive buffering mechanism is used to generate a smooth control instruction set;

[0065] It should be noted that the "gain adaptive buffering mechanism" refers to the following: after feature handover is completed, dynamic analysis is performed on the target position, velocity vector, attitude change information, etc. contained in the unified perception state set. Based on factors such as the target state change rate, directional fluctuation amplitude, and command update frequency, the control gain coefficient is adaptively adjusted. On this basis, a buffer time window is introduced to construct a control command generation strategy with time smoothing and amplitude suppression characteristics.

[0066] Understandably, the design of this mechanism is essentially aimed at suppressing the common problems of "command abrupt changes" and "feedback oscillations" in multi-UAV cooperative control. Especially when the target state undergoes abrupt changes or the perception boundary shows a fusion transition, it can automatically reduce the gain to avoid control overshoot, while delaying the output of non-critical commands to achieve the dynamic control characteristics of "flexible takeover." The smooth control command set helps improve the stability of trajectory tracking in multi-UAV formations, reduces disturbances during attitude adjustment, and provides dynamic support for downstream path planning and attitude control.

[0067] It should be understood that, compared with traditional fixed-gain control mechanisms, fixed PID gain coefficients are prone to controller oscillations and path deviations when faced with drastic changes in perceived input. This invention, by introducing state-aware driven gain adjustment and buffer filtering processes, achieves a better trade-off between response speed and smoothness in control commands, effectively improving the reliability of control commands in complex scenarios (such as high-frequency obstacle crossing and rapid target infiltration).

[0068] For example, in a scenario test, after the lead drone and the relay drone completed the target perception handover, the target suddenly veered to the left and increased its flight speed. The high rate of change of speed recorded in the perception state set triggered a gain adaptive buffer mechanism to activate a low-gain interval. Within this buffer, the control commands were weighted and averaged across multiple points within a sliding window, outputting a command sequence that smoothly transitioned from the original "sharp turn and sudden braking" commands to "gradual turn and uniform deceleration." This ensured that the relay drone smoothly took over the control task at the path handover point without any sudden oscillations.

[0069] Step S40: Based on the smooth control instruction set, a limiting and suppression mechanism based on rate constraints is used to perform key control quantity limiting tasks, and the limiting and smooth control instruction set is output.

[0070] It should be noted that the "variable rate constraint-based amplitude limiting and suppression mechanism" refers to: on the basis of the generated smooth control command set, further apply dynamic amplitude limiting control based on first-order variable rate constraints to the key control quantities involved (such as acceleration command, yaw rate, thrust change rate, etc.) to ensure that they are within the physical safety boundary.

[0071] It should be understood that, compared with traditional clipping mechanisms (such as simple clip operations), the "variable rate constraint clipping" introduced in this invention is more dynamic and adaptive. Traditional methods often directly truncate control values ​​with fixed upper and lower limits, which may lead to sudden changes in commands, delayed or discontinuous target responses; while this invention monitors and suppresses the change amplitude of two consecutive cycles, achieving gradual limitation at the "rate of change" level, avoiding discontinuous jump behavior, and making the control output have stronger time consistency and physical constraint adaptation capabilities, thus exhibiting higher reliability and execution smoothness in application scenarios such as UAV cooperative control and trajectory following.

[0072] Step S50: Apply the amplitude-limiting smooth control instruction set to perform adaptive adjustment of the driving vehicle platform.

[0073] It should be noted that this step is the final execution link of the entire control chain. "Adaptive adjustment of the driving vehicle platform" refers to the control of the unmanned vehicle platform to perform actions such as attitude adjustment, trajectory following, and power output adjustment based on the amplitude limiting smooth control instruction set output in step S40, and to dynamically correct the control channel output based on real-time sensor feedback, forming a closed-loop adaptive control mechanism.

[0074] Example 2: Furthermore, the present invention provides an adaptive adjustment system for a driving platform under UAV collaboration, employing an adaptive adjustment method for a driving platform under UAV collaboration as described in the above embodiments, which can solve the technical problem of adaptive adjustment of a driving platform under UAV collaboration. Compared with the prior art, the beneficial effects of the adaptive adjustment system for a driving platform under UAV collaboration provided by the present invention are the same as the beneficial effects of the adaptive adjustment method for a driving platform under UAV collaboration provided in the above embodiments, and other technical features of the adaptive adjustment system for a driving platform under UAV collaboration are the same as those disclosed in the methods of the above embodiments, and will not be repeated here.

[0075] Example 3: This invention provides an adaptive adjustment device for a drone-assisted driving platform. Referring to Figure 3, the drone-assisted adaptive adjustment device includes: at least one processor; and a memory communicatively connected to the at least one processor. The memory stores instructions executable by the at least one processor, which, when executed, enable the at least one processor to perform the drone-assisted adaptive adjustment method for a driving platform as described in Example 1. The drone-assisted adaptive adjustment device in this embodiment may include, but is not limited to, mobile terminals such as mobile phones, laptops, digital radio receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Description), PMPs (Portable Media Players), and in-vehicle terminals (e.g., in-vehicle navigation terminals), as well as fixed terminals such as digital TVs and desktop computers. This drone-assisted adaptive adjustment device is merely an example and should not limit the functionality or scope of the invention. An adaptive adjustment device for a drone-cooperative driving platform may include a processing unit 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory 1002 or a program loaded from a storage device 1003 into a random access memory 1004. The random access memory 1004 also stores various programs and data required for the operation of the drone-cooperative adaptive adjustment device. The processing unit 1001, the read-only memory 1002, and the random access memory 1004 are interconnected via a bus 1005. An I / O interface 1006 is also connected to the bus. Typically, the following systems can be connected to the I / O interface 1006: input devices 1007 including, for example, a touchscreen, touchpad, keyboard, mouse, image sensor, microphone, accelerometer, gyroscope, etc.; output devices 1008 including, for example, a liquid crystal display (LCD), speaker, vibrator, etc.; storage devices 1003 including, for example, magnetic tape, hard disk, etc.; and communication devices 1009. The communication device 1009 allows a drone-cooperative adaptive adjustment device for a driving platform to communicate wirelessly or wiredly with other devices to exchange data. While the figure shows a drone-cooperative adaptive adjustment device for a driving platform with various systems, it should be understood that implementation of or possession of all the systems shown is not required. More or fewer systems may be implemented alternatively.

[0076] Example 4: This invention also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the above-described adaptive adjustment method for a drone-assisted driving platform. The computer program product provided by this invention can solve the technical problem of adaptive adjustment of a drone-assisted driving platform. Compared with the prior art, the beneficial effects of the computer program product provided by this invention are the same as those of the adaptive adjustment method for a drone-assisted driving platform provided in the above embodiments, and will not be repeated here.

[0077] In particular, according to the embodiments disclosed in this invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this invention include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from storage device 1003, or installed from read-only memory 1002. When the computer program is executed by processing device 1001, it performs the functions defined in the methods of the embodiments disclosed in this invention.

[0078] It should be understood that the various parts disclosed in this invention can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any suitable manner in one or more embodiments or examples.

[0079] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. An adaptive adjustment method for a driving and transport platform in collaboration with unmanned aerial vehicles (UAVs), characterized in that, The method includes: Step S10: Obtaining target transportation route information and site topology information, and determining based on the target transportation route information and site topology information. The flight path set of the collaborative UAV is used to perform the field-of-view overlap determination task based on the flight path set and the relay redundancy mapping mechanism based on the residual task window constraint, and outputs the field-of-view overlap starting region set F; Step S20: Based on the field-of-view overlap starting region set F, the feature handover task is performed based on the dual-source synchronous locking and weighted progressive fusion mechanism, and the unified perception state set is output. Among them, based on the set of overlapping field-of-view regions F, a feature handover task is performed using a dual-source synchronous locking and weighted progressive fusion mechanism, outputting a unified perception state set. The steps specifically include: obtaining the broadcast vehicle status set sent by the current leading UAV b to the relay UAV m based on the set of overlapping field-of-view starting regions F. ， ， For location information, For speed information, For heading angle information; the relay drone m based on the broadcast vehicle status set Perform vehicle recognition and target locking tasks, and output local feedback perception results. Based on local feedback perception results and broadcast vehicle status set A linear interpolation fusion method based on time-progressive smooth collaboration is used to obtain a unified perception state result. Step S30: Based on the unified sensing state set A gain-adaptive buffering mechanism is employed to generate a smooth control instruction set; among which, a unified sensing state set is used. The steps for generating a smooth control instruction set using a gain adaptive buffering mechanism specifically include: based on a unified sensing state set. Generate initial control commands at any time t Simultaneously acquire historical control commands at time t-1. Based on initial control commands and historical control commands The current smoothing control command is constructed using a first-order IIR filter structure with weighted dynamic suppression. Current smooth control command The formula is expressed as: in, The introduced dynamic gain suppression factor, adjusted according to the drone relay status, is defined as follows: in, Basic gain suppression factor, ； This is the gain adjustment intensity coefficient. Step S40: Based on the smooth control instruction set, a rate-constrained limiting and suppression mechanism is used to execute the key control quantity limiting task, and the rate-constrained smooth control instruction set is output. Specifically, this step includes: obtaining the current steering angle control instruction and the current acceleration control instruction from the smooth control instruction set, and obtaining the preset desired steering angle control instruction and desired acceleration control instruction; constructing the first control quantity rate factor based on the current steering angle control instruction and the desired steering angle control instruction using residual analysis and normalization calculation methods. Based on the current acceleration control command and the desired acceleration control command, the second control variable rate factor is constructed using residual analysis and normalization calculation methods. Based on the first control variable rate factor Second control quantity rate factor The steering angle variation limiting threshold is constructed using the linear annealing mapping method. and acceleration rate limiting threshold Based on steering angle variation limiting threshold and acceleration rate limiting threshold The smooth control instruction set is compressed using a bilateral symmetric amplitude limiting function, and the final output amplitude limiting smooth control instruction set is obtained; Step S50: Apply the amplitude limiting smooth control instruction set to perform adaptive adjustment of the driving vehicle platform.

2. The adaptive adjustment method for a driving and transport platform under UAV collaboration as described in claim 1, characterized in that, In step S10, target transportation route information and site topology information are obtained, and based on the target transportation route information and site topology information, the following is determined: The steps involved are: Step S101: Obtaining target transportation route information and site topology information, and determining the field of view overlap determination task based on the target transportation route information and site topology information. The set of flight paths of a collaborative UAV; obtain the first flight path in the set of flight paths. Remaining battery information for the drone Unit power consumption information Airspace occupancy permit time information Information on the estimated completion time of the current task Based on remaining battery power information Unit power consumption information Airspace occupancy permit time information Information on the estimated completion time of the current task The first method is constructed using a joint decision-making approach based on energy consumption feasible region constraints and space occupancy restrictions. Remaining mission window for drones Remaining task window Used to characterize the redundant time range within which the UAV can continue to be used for relay observation after completing its main task; Step S102: Based on the remaining task window Constructing candidate relay periods ， In order to be with the first The start time of the candidate relay period for the j-th drone adjacent to the first drone; In order to be with the first The candidate relay time end time of the j-th drone adjacent to the first drone; ， Step S103: Jointly with the first Flight altitude of the drone With depression angle Determine the ground field of view coverage radius United Nations Flight altitude of the drone With depression angle Determine the ground field of view coverage radius ; Get the The drone and the first The horizontal distance between the two drones on the ground If satisfied Then determine the candidate relay period. For the effective relay period, the effective relay period is regarded as the starting region of field of view overlap, and the final output is the set of starting regions of field of view overlap F.

3. The adaptive adjustment method for a driving and transport platform under UAV collaboration as described in claim 1, characterized in that, In step S20, the unified perception state result is obtained. The formula is expressed as: in, Let be the fusion weight function at any time t. ， This is the moment of relay switching; This is the effective relay period.

4. The adaptive adjustment method for a driving and transport platform under UAV collaboration as described in claim 3, characterized in that, In step S40, the formula for the bilateral symmetric limiting function is expressed as follows: in, Let be a bilaterally symmetric amplitude-limiting function at any time t; For smooth control of any instruction signal in the instruction set The original value at any time t; For smooth control of any instruction signal in the instruction set The control value at the previous time t-1; For command signals The rate limiting threshold is constructed at any time t, including the steering angle rate limiting threshold. and acceleration rate limiting threshold ； It is a symbolic function.

5. An adaptive adjustment system for a driving platform in collaboration with unmanned aerial vehicles (UAVs), applied to the adaptive adjustment method for a driving platform in collaboration with UAVs as described in any one of claims 1 to 4, characterized in that, The UAV-assisted adaptive adjustment system for the driving and transportation platform includes: a path planning module, used to acquire target transportation route information and site topology information, and to determine the path based on the target transportation route information and site topology information. The system uses a set of flight paths from collaborative UAVs. Based on this set, a relay redundancy mapping mechanism with residual task window constraints is employed to perform a field-of-view overlap determination task, outputting a set of field-of-view overlap initiation regions, F. A state handover module is used to perform a feature handover task based on the field-of-view overlap initiation region set F, employing a dual-source synchronous locking and weighted progressive fusion mechanism, outputting a unified perception state set. Among them, based on the set of overlapping field-of-view regions F, a feature handover task is performed using a dual-source synchronous locking and weighted progressive fusion mechanism, outputting a unified perception state set. The steps specifically include: obtaining the broadcast vehicle status set sent by the current leading UAV b to the relay UAV m based on the set of overlapping field-of-view starting regions F. ， ， For location information, For speed information, For heading angle information; the relay drone m based on the broadcast vehicle status set Perform vehicle recognition and target locking tasks, and output local feedback perception results. Based on local feedback perception results and broadcast vehicle status set A linear interpolation fusion method based on time-progressive smooth collaboration is used to obtain a unified perception state result. Instruction smoothing module, used for processing based on unified sensing state set. A gain-adaptive buffering mechanism is employed to generate a smooth control instruction set; among which, a unified sensing state set is used. The steps for generating a smooth control instruction set using a gain adaptive buffering mechanism specifically include: based on a unified sensing state set. Generate initial control commands at any time t Simultaneously acquire historical control commands at time t-1. Based on initial control commands and historical control commands The current smoothing control command is constructed using a first-order IIR filter structure with weighted dynamic suppression. Current smooth control command The formula is expressed as: in, The introduced dynamic gain suppression factor, adjusted according to the drone relay status, is defined as follows: in, Basic gain suppression factor, ； This is the gain adjustment intensity coefficient. The amplitude limiting and suppression module is used to execute key control quantity limiting tasks based on a rate constraint-based amplitude limiting and suppression mechanism using a smooth control instruction set, and outputs an amplitude limiting and smooth control instruction set. Specifically, the steps of executing key control quantity limiting tasks based on a rate constraint-based amplitude limiting and suppression mechanism using a smooth control instruction set and outputting the amplitude limiting and smooth control instruction set include: obtaining the current steering angle control instruction and the current acceleration control instruction from the smooth control instruction set, and obtaining preset desired steering angle control instructions and desired acceleration control instructions; and constructing a first control quantity rate factor based on the current steering angle control instruction and the desired steering angle control instruction using residual analysis and normalization calculation methods. Based on the current acceleration control command and the desired acceleration control command, the second control variable rate factor is constructed using residual analysis and normalization calculation methods. Based on the first control variable rate factor Second control quantity rate factor The steering angle variation limiting threshold is constructed using the linear annealing mapping method. and acceleration rate limiting threshold Based on steering angle variation limiting threshold and acceleration rate limiting threshold A bilateral symmetric amplitude limiting function is used to compress the smooth control instruction set, and the final output amplitude limiting smooth control instruction set is obtained; the platform execution module is used to apply the amplitude limiting smooth control instruction set to perform adaptive adjustment of the driving vehicle platform.

6. An adaptive adjustment device for a driving and transport platform in collaboration with unmanned aerial vehicles (UAVs), characterized in that, The drone-assisted driving platform adaptive adjustment device includes: a memory, a processor, and a drone-assisted driving platform adaptive adjustment program stored in the memory and executable on the processor. When the drone-assisted driving platform adaptive adjustment program is executed by the processor, it implements a drone-assisted driving platform adaptive adjustment method according to any one of claims 1 to 4.

7. A computer program product, characterized in that, The computer program product includes an adaptive adjustment program for a driving platform under UAV collaboration, which, when executed by a processor, implements an adaptive adjustment method for a driving platform under UAV collaboration as described in any one of claims 1 to 4.

Images