CN-122008186-A - Near-ground target grabbing planning method for unmanned aerial vehicle with arm based on hawk falcon hunting swing arm behaviors

CN122008186ACN 122008186 ACN122008186 ACN 122008186ACN-122008186-A

Abstract

The invention discloses a near-ground target grabbing planning method of an unmanned aerial vehicle with arms based on hawk hunting swing arm behaviors, which comprises the steps of firstly, carrying out hawk hunting analysis and modeling of an unmanned aerial vehicle with arms, dividing a grabbing process into three stages of before grabbing, during grabbing and after grabbing, and establishing unmanned aerial vehicle motion constraint according to hawk behavior characteristics of each stage, secondly, designing grabbing planning optimization problems with energy optimization as targets and introducing a position relaxation factor to adjust arrival constraint, thirdly, establishing planning constraint of each stage of hawk approaching, grabbing and returning swing arm behaviors, fourthly, constructing dynamics planning constraint of an unmanned aerial vehicle system and a mechanical arm system, fifthly, solving grabbing optimization problems and generating a track, and sixthly, outputting grabbing planning tracks conforming to hawk hunting swing arm behaviors. According to the invention, through the energy optimization target and the relaxation factor design, the feasibility and the instantaneity of solving are enhanced while the physical constraint of the system is met, and the engineering realizability is good.

Inventors

DUAN HAIBIN
WU TONGYAN
ZHANG ZHAOYU
SUN YONGBIN

Assignees

北京航空航天大学

Dates

Publication Date: 20260512
Application Date: 20251230

Claims (4)

1. A near-ground target grabbing planning method of an unmanned aerial vehicle with arms based on hawk and falcon hunting swing arm behaviors is characterized by comprising the following steps: Step one, hawk falcon hunting analysis and armful unmanned aerial vehicle system modeling S11, analyzing the behavior of a hawk falcon hunting swing arm; dividing a hawk hunting swing arm-based process into three stages of before, during and after grabbing, establishing optimization constraint conforming to a hawk behavior characteristic of the unmanned aerial vehicle with an arm according to each stage, and realizing track generation according to the energy design optimization problem; S12, establishing a basic model of the unmanned aerial vehicle with the arm; s13, carrying out task grabbing stage division on the unmanned aerial vehicle with the arm; step two, designing optimization problem of near-ground target grabbing planning of unmanned aerial vehicle with arm Converting the guiding planning problem into a planning optimization solving problem, establishing an optimization target through an energy relation, and simultaneously ensuring that the inherent physical constraint of the unmanned aerial vehicle and the behavior characteristic constraint of the hawk swing arm are met before, during and after grabbing; s21, designing a near-to-ground target grabbing optimization problem; S22, designing a position relaxation factor; Step three, planning constraint of designing simulated hawk falcon hunting swing arm behavior mapping S31, planning and restraining a hawk-like swing arm unmanned aerial vehicle with arms at stage 0; S32, planning and restraining the unmanned aerial vehicle with the arm, which is used for imitating hawk and grabbing a swing arm in the stage 1; S33, planning and restraining the hawk-like return swing arm with arm unmanned aerial vehicle in the stage 2; Step four, power planning constraint of unmanned aerial vehicle system with arm S41, designing kinematic constraint of the unmanned aerial vehicle system, and limiting amplitude of the speed and the acceleration of an expected instruction to ensure that the unmanned aerial vehicle can normally execute; s42, performing mechanical arm system kinematics constraint design, and specifically performing constraint design on the executable angle of the mechanical arm joint; solving the ground-proximity target grabbing optimization problem of the unmanned aerial vehicle with the arm and generating a track Obtaining a final planning result by solving an optimization problem, and requiring a solver to carry out algorithm solving aiming at an optimization target and hawk and falcon hunting swing arm optimization constraint, and ensuring normal execution of solving by adjusting a relaxation factor when an optimization solving contradiction exists; Step six, outputting a grabbing planning track based on hawk falcon hunting swing arm behaviors The method comprises the steps of providing expected positions and expected speeds of unmanned aerial vehicles with arms and expected angles and expected angular speeds of mechanical arms at each moment based on optimized state variables, ensuring that controllers of the unmanned aerial vehicles with arms can execute expected instructions, and finally achieving grabbing planning tracks based on hawk hunting swing arm behaviors.
2. The method for planning the near-ground target grabbing of the unmanned aerial vehicle with the arm based on hawk and falcon swing arm behaviors of claim 1 is characterized in that the steps of grabbing task phases of the unmanned aerial vehicle with the arm are divided into three phases of approaching, grabbing target phases and grabbing carrying flying of the unmanned aerial vehicle with the arm, and the three phases are phase 0, phase 1 and phase 2.
3. The near-ground target grabbing planning method of the unmanned aerial vehicle with the arm based on hawk and falcon hunting swing arm behaviors of claim 1 is characterized by comprising the following specific process of: s21, designing ground-near target grabbing optimization problem The grabbing design optimization problem for the unmanned aerial vehicle with the arm is as follows: (1) Wherein, the As a function of the cost function value, In order to achieve a total execution time, the system, The motion vector of the unmanned aerial vehicle system, , For the angular movement speed of the mechanical arm, the length of the first section of mechanical arm connected with the unmanned aerial vehicle is as follows The length of the second section of mechanical arm is l 2 , , , , Are all the relaxation factors, and the preparation method is simple, The relaxation factor is used for restraining the belt arm unmanned aerial vehicle to grasp to reach the target constraint; s22, site relaxation factor design In the specific optimization arrival design, a relaxation factor for reducing the constraint limit of the arrival condition is designed, and the grabbing target position when the stage 0 is completed is firstly set as The tail end position of the mechanical arm of the unmanned aerial vehicle with the arm when the stage 2 is completed is On this basis, the relaxation factor is defined as: (3) Wherein, the Is a position vector of the unmanned aerial vehicle with an arm, The position of the armed drone at the moment of performing the gripping for phase 0, And (3) after the stage 2 is grabbed, the tail end position of the unmanned aerial vehicle with the arm at the time of returning is completed, and the constraint on the arrival position is relaxed by setting a relaxation factor and adjusting the relaxation factor.
4. The method for controlling the fight against the unmanned aerial vehicle group based on the eagle pigeon game according to claim 1, wherein the specific process of the third step is as follows: s31, planning constraint of hawk-like hawk approaching swing arm phase 0 unmanned aerial vehicle with arm At the end time of the stage 0, the tail end speed of the unmanned aerial vehicle with the arm is required to have no vertical direction speed, and the tail end direction of the mechanical arm is required to be in the horizontal direction in the mechanical arm part, so that the motion constraint is established as follows: (4) Wherein, the Aiming at a desired angle for the hawk falcon hunting swing arm, the hawk falcon hunting swing arm is set as , , , For smaller positive error boundaries, the determination is made in terms of actual conditions and solver performance conditions, , For the target speed, set to 0, As a factor of the angular relaxation, The arm swing angle at which the gripping moment is performed for phase 0, 、、 For the x, y and z direction speeds of the tail end of the mechanical arm at the stage 0 executing grabbing moment, the unmanned aerial vehicle with the arm is required to ensure the speed consistent with the target at the stage 0 completing moment, and alignment is carried out according to the swing arm characteristics of the hawk hunting preparation stage, and in addition, the stage constraint simultaneously comprises the formula ; S32, planning constraint of hawk-simulated hawk grabbing swing arm with arm unmanned aerial vehicle in stage 1 The restraint of the hawk-like grabbing swing arm is planned and designed in the stage 1 as follows: (5) Wherein, the Aiming at a desired angle for the hawk falcon hunting swing arm, the hawk falcon hunting swing arm is set as , For smaller positive error boundaries, the determination is made in terms of actual conditions and solver performance conditions, As a factor of the angular relaxation, For the arm angle after the swing arm is completed in stage 1, The end velocity vector when the grab is performed for phase 0, The tail end speed vector after the swing arm is finished in the stage 1 is considered to finish target grabbing, and the swing arm is carried out by ensuring the speed to be unchanged, so that the system adjusts the mechanical arm to be more in line with the carrying flying state in the dynamic movement process, and the carrying target flying efficiency is improved; S33, planning constraint of hawk falcon simulated hawk return swing arm phase 2 unmanned aerial vehicle with arm The planning constraints of the phase 2 design system are: (6) Wherein, the , , For smaller positive error boundaries, the determination is made in terms of actual conditions and solver performance conditions, , For target speed, the constraint contains the formula at the same time ; 、、 The angle design is adjusted on the basis of the angle of the stage 1, the angle is optimized only through energy cost without specific constraint on the angle, and the energy is ensured to be optimal in the dynamic process.

Description

Near-ground target grabbing planning method for unmanned aerial vehicle with arm based on hawk falcon hunting swing arm behaviors Technical Field The invention relates to a near-ground target grabbing planning method of an unmanned aerial vehicle with arms based on hawk falcon hunting swing arm behaviors, and belongs to the fields of grabbing design and task planning of unmanned aerial vehicles with arms. Background With the rapid development of the intelligent and autonomous unmanned technology, the low-altitude economy has become the research focus of the current unmanned technology. Unmanned aerial vehicles have been widely used in various task scenarios such as military reconnaissance, environmental monitoring, facility inspection, etc. In performing these typical tasks, the drone typically selects a near-ground or ground target as the work object. The task execution is realized through the process of quickly and safely descending from a high-altitude cruising area to a target near-airspace, and then climbing back to the high altitude to continue flying after the specific task is executed. The unmanned aerial vehicle with the arm is used as a novel unmanned aerial vehicle type at present, the action execution is rich by virtue of the unmanned aerial vehicle type, the adaptation scene is wide, and the unmanned aerial vehicle type is becoming a research hotspot in the field of the current unmanned aerial vehicle. However, its maneuverability requires consideration of the robotic arm and drone synergy, resulting in inherent constraints on flight control. Therefore, when performing tasks requiring access to the ground, special guidance algorithms must be designed to ensure that the entire task process is completed safely and efficiently. At present, guidance algorithm research for the unmanned aerial vehicle with the arm mainly focuses on cooperative control or tracking and grabbing tasks of specific targets. The method comprises the steps of generating a formation guiding method of a speed and course instruction through design and combining research works such as vector field guiding and model predictive control so as to solve the problems of air obstacle avoidance, ground target tracking and the like. These studies are mostly directed to general guidance scenes, and lack of targeted design of two phases of "near-ground target approach" and "return" with different guidance targets and status features results in limited effectiveness in near-ground guidance tasks. In addition, the existing guidance rules are designed based on custom rules, and cannot be systematically analyzed and designed from the aspects of system self-limitation and energy optimization. Therefore, it is necessary to design a guiding algorithm specially used for the fixed-wing unmanned aerial vehicle to perform the near-ground target task, which is important for expanding the effective application of the fixed-wing unmanned aerial vehicle in multiple fields. In nature, hawks and other birds exhibit excellent maneuverability and environmental adaptability during hunting. The device can flexibly adjust the dive attack track according to the dynamics of the prey, and finish high-speed approaching, accurate guidance and dynamic obstacle avoidance in a near energy optimal mode. Studies have shown that hawk has a "delayed guidance" strategy when approaching targets, where steering regulation presents a slow to fast process, rather than direct impact on the prey. Research on Harris eagle and other species further reveals its group collaborative hunting mechanisms, including a series of behavioral patterns of collaborative search, localization and attack, which have become important biomimetic sources for heuristic optimization algorithms. The whole process that hawk is returned to the air after the hawk is pushed down from the high altitude to catch ground targets has high similarity with the process that the unmanned plane executes ground-approaching tasks and returns to the air. The mapping between the swing arm behaviors and the near-ground target grabbing tasks of the unmanned aerial vehicle with the arm in the hawk and falcon hunting process is established, so that theoretical support and interpretability can be provided for the unmanned aerial vehicle to efficiently execute the near-ground tasks. In summary, the invention provides a near-ground target grabbing planning method for an unmanned aerial vehicle with arms based on hawk falcon hunting arm behaviors, which is characterized in that an action constraint relation is established through model analysis in the hawk falcon hunting process, and the hawk falcon hunting arm behavior model is mapped to an approach-grabbing-returning process of the unmanned aerial vehicle with arms to perform near-ground grabbing targets, so that the planning method accords with power constraint, is simple and efficient, has good executable of a generated result and has a certain practical application value.