CN116518974A - Conflict-free route planning method based on airspace grids - Google Patents
Conflict-free route planning method based on airspace grids Download PDFInfo
- Publication number
- CN116518974A CN116518974A CN202310477933.9A CN202310477933A CN116518974A CN 116518974 A CN116518974 A CN 116518974A CN 202310477933 A CN202310477933 A CN 202310477933A CN 116518974 A CN116518974 A CN 116518974A
- Authority
- CN
- China
- Prior art keywords
- route
- path
- time
- airspace
- aircraft
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 41
- 238000009499 grossing Methods 0.000 claims abstract description 33
- 230000002787 reinforcement Effects 0.000 claims abstract description 8
- 230000006870 function Effects 0.000 claims description 20
- 230000008569 process Effects 0.000 claims description 9
- 230000009471 action Effects 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 5
- 230000003068 static effect Effects 0.000 claims description 5
- 238000010276 construction Methods 0.000 claims description 3
- 238000013507 mapping Methods 0.000 claims description 3
- 230000008859 change Effects 0.000 description 4
- 238000013528 artificial neural network Methods 0.000 description 3
- 230000004888 barrier function Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 230000007704 transition Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 210000005036 nerve Anatomy 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/20—Instruments for performing navigational calculations
Landscapes
- Engineering & Computer Science (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Automation & Control Theory (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention provides a conflict-free route planning method based on airspace grids, and relates to the field of path planning; the method comprises the following steps: firstly, an aircraft deploys an airborne platform, four-dimensional track data are collected, and flight starting point, end point and obstacle information are sent to a route planning module; the route planning module constructs a four-dimensional space-time airspace grid according to the coordinate relation of different three-dimensional spaces on each time slice. Then, on the basis of the airspace grids, a four-dimensional space-time traffic value graph is calculated according to reinforcement learning iteration; selecting a conflict-free path with the minimum cost from the traffic value graph, and conveying the conflict-free path to a path smoothing module; and smoothing the route control point sequence according to the physical characteristics and the dynamic characteristics of the aircraft by a route smoothing algorithm, and finally sending a smoothed result to an airborne platform, wherein the airborne platform controls the flight of the aircraft according to the route. The invention derives the traffic value graph through the value iterative network, thereby greatly reducing the running time of the system and improving the effectiveness.
Description
Technical Field
The invention relates to the field of path planning, in particular to a conflict-free route planning method based on airspace grids.
Background
The route planning means: by using various path planning algorithms, a short-distance, less time-consuming and collision-free flight path is planned for the aircraft after the start point and the end point are given. Traditional route planning methods such as Dijiestra algorithm, ant colony algorithm or A-type algorithm focus on planning the optimal path with the shortest flight distance and the least time consumption; however, when there are many obstacles in the space domain and the obstacles dynamically change with time, the path planning algorithm described above is not very good and may even plan a path with collision; and the importance of path smoothing on the route planning is not considered in the traditional route planning method, and when the route has too sharp corners, the riding experience of personnel in an aircraft can be seriously influenced, and even the air traffic safety is influenced. Therefore, it is necessary to invent a route plan that can be adapted to the dynamic changes of obstacles and that can be planned to be smooth and collision free.
In the prior art, as the low-altitude rescue aircraft track planning method based on the three-dimensional airspace grid with the application number of 20151250418. X, the conflict-free track planning of multiple aircraft under the condition of low-altitude complex terrain and weather distribution and consideration of the performance of the aircraft under the aviation rescue is solved, and a basis is provided for the aircraft flight planning of an aviation emergency rescue command system. However, when the method is used for planning a path, the constraint factors are considered to only comprise fixed obstacles such as terrain, weather and the like, and the factors of dynamic changes of the obstacles along with time are not considered; i.e. path planning is only performed in the topology on the static map of the obstacle location at the current moment. When the position of the obstacle changes at the next moment, the planning result of the upper path fails and the planning needs to be carried out again.
The application number is 201710255262.6, namely a multi-target path planning joint search method for unmanned aerial vehicles in a low-altitude urban environment, firstly, a path with the minimum cost from a starting point to an ending point is planned off-line, and then, when an unknown obstacle is detected, a changed track is planned on-line by adopting on-line search so as to avoid dynamic obstacles. The online search has smaller search space, and can rapidly and online re-plan a section of safe path for the unmanned aerial vehicle, thereby meeting the requirement of the unmanned aerial vehicle on the real-time performance of path planning. Although the influence of the dynamic obstacle is considered, a new path is planned by going to the online search when the dynamic obstacle is detected in real time, the requirement on real-time performance is greatly reduced, and the optimal path is not considered from the global point of view. The path planning for dynamically changing obstacles is to re-plan a new path on line after detecting the change. Therefore, the path planning result of the technology is not a globally optimal path, and even unnecessary start-stop and avoidance operations can be caused, so that the driving efficiency and the riding experience are reduced.
The cross-domain heterogeneous cluster path planning method based on reinforcement learning has the advantages that a Markov decision process and a reward function of cross-domain heterogeneous cluster maneuver are constructed, and a MADDPG algorithm is utilized to solve individual maneuver strategies of clusters, so that the path planning of the cross-domain heterogeneous clusters is realized. The path planning can be realized by the method provided by the invention as long as the clustered individuals have relevant sensors. Under the method, different aircrafts from airspace, sea area, land area and other areas can jointly carry out path planning, so that 'cross-domain heterogeneous path planning' is realized. However, the method adopts the fully-connected neural network to carry out path planning, so that the model needs to be retrained according to different application scenes, and the method cannot adapt to an environment with higher real-time requirements.
Disclosure of Invention
In order to solve the problems, the invention discloses a conflict-free route planning method based on a space grid, which comprises the steps of constructing a space grid of four-dimensional space and time by increasing a time dimension on the basis of a three-dimensional space after receiving data transmitted by an airborne platform, calculating a four-dimensional space and time passing value graph, constructing an optimal conflict-free path according to the value graph, and finally outputting a final route planning result after processing by a path smoothing system. According to the invention, the obstacle information of each time step is comprehensively considered, the value iteration is carried out on four-dimensional space time to obtain the globally optimal collision-free path in a low-cost mode, and the path smoothing process is added, so that the flight track of the aircraft is more in line with the physical characteristics and the dynamic characteristics.
The method comprises the following specific steps:
firstly, an aircraft deploys an airborne platform, four-dimensional track data are collected, flight starting point, end point and obstacle information are processed into a dynamic point track sequence, and the dynamic point track sequence is sent to a route planning module;
and step two, constructing a four-dimensional space-time airspace grid by the route planning module according to the coordinate relation of different three-dimensional spaces on each time slice.
The specific construction process is as follows:
and generating a three-dimensional target airspace by taking a starting point and an ending point of the aircraft as boundaries, dividing the three-dimensional target airspace into a plurality of four-dimensional grids, setting the length, the width and the height of each four-dimensional grid unit according to the standard interval distance of the aircraft, and corresponding the positions of the flight path and the obstacle of the aircraft to the set four-dimensional grids according to time slices.
Thirdly, based on the airspace grid, calculating a four-dimensional space-time passing value graph according to reinforcement learning iteration; and selecting a conflict-free path with the minimum cost from the traffic value graph, and conveying the conflict-free path to the path smoothing module.
The specific process is as follows:
step 301, a cost function v(s) and a return function r(s) of a defined state s;
the cost function v(s) is defined as: the expectation of rewards earned by the aircraft for reaching the target location along a path selected using strategy pi, starting from state s; policy pi is defined as the mapping of state s to the probability of any feasible action a;
the definition of the return function r(s) is:
wherein S is obstacle Occupied grid for other elements (barriers) in the environment other than the current agentA set of locations; s is S goal Grid location for the target;
step 302, traversing the paths of the aircraft from each position to the destination according to the four-dimensional space-time airspace grid map, calculating the return, namely the passing cost, of each position by using a return function, and storing the return into a return map;
for any state s= (t, i, j, k), the values in the reward graph represent the reward values given when time t is at grid position (i, j, k).
Step 303, calculating returns at all moments based on the return graphs by using a value iterative algorithm, and calculating a four-dimensional space-time passing value graph converged at the moment T through multiple iterations;
firstly, in the initialization stage of an iterative algorithm, setting the value of all positions except the target position at all times to be- ≡, and setting the value of the target position at all times to be 0; n is the maximum input frame number of the value iterative algorithm.
Then, during calculation, a three-dimensional space value iterative algorithm is used for calculating a traffic value graph at the moment T by using a grid map of the last frame in a static map mode;
and after the traffic value map at the last moment T converges, sequentially calculating by a four-dimensional space-time value iteration method to obtain traffic value maps at the moment T-1, T-2, … and 0, and finally updating the traffic values at all positions.
The space-time cost function is updated according to the following formula:
step 304, when the value of each state in the traffic value graph is the optimal value, the selected optimal flight route path is the optimal flight route path;
the path is a flight route control point sequence composed of grid coordinates.
And step four, after the route smoothing module obtains a conflict-free route with the minimum cost, a route smoothing algorithm smoothes a route control point sequence according to physical characteristics and dynamic characteristics of the aircraft, and finally, the smoothed result is sent to an airborne platform, and the airborne platform controls the flight of the aircraft according to the route.
The data smoothing adopts a linear interpolation method.
The invention has the advantages that:
1) Compared with the prior art that path planning is carried out only by means of the topological structure on the static map of the obstacle position at the current moment, the conflict-free route planning method based on the airspace grids can be suitable for scenes of dynamic changes of the obstacle by constructing the four-dimensional airspace grids and the four-dimensional spatio-temporal value map, and an optimal conflict-free path can be planned when the obstacle changes dynamically.
2) Compared with the prior art that only reinforcement learning and a neural network are used for path planning, the conflict-free route planning method based on the airspace grids is used for deducing the traffic value graph through the value iterative network, avoids a long model training process, greatly reduces the system running time and improves the effectiveness.
3) The conflict-free route planning method based on the airspace grids also utilizes a path smoothing algorithm to smooth the route control point sequence according to the physical characteristics and the dynamic characteristics of the aircraft, and ensures that the finally planned route is stable and reliable.
Drawings
FIG. 1 is a schematic diagram of a system for planning a route in accordance with the present invention;
FIG. 2 is a flow chart of a collision-free route planning method based on airspace grids.
Detailed Description
The present invention is further described in detail below with reference to the drawings and examples for the purpose of facilitating understanding and practicing the present invention by those of ordinary skill in the art. It is obvious that the described embodiments are only some embodiments of the present invention, but not all embodiments, and especially the present invention does not limit the type of intelligent optimization and conventional optimization algorithm of the group in the technical solution. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present invention without making any inventive effort, shall fall within the scope of the present invention.
The invention discloses a conflict-free route planning method based on airspace grids, which is based on the system architecture of a conflict-free route planning system of airspace grids, as shown in figure 1, and comprises the following steps: an airborne platform, a path planning system and a path smoothing system:
1) Airborne platform
The aircraft-mounted platform deployed on the aircraft can send information such as a starting point, a finishing point, obstacles and the like to the route planning module, the route planning module plans a collision-free route after calculation processing, and the aircraft-mounted platform can select to control the aircraft to fly according to the route data.
2) Path planning system
After obtaining the starting point, the end point and the dynamic barrier information, the path planning system constructs a four-dimensional space-time airspace grid according to the coordinate relation of different three-dimensional spaces on each time slice; and continuously and iteratively calculating a traffic value graph on the basis of the airspace grids, wherein the traffic value graph reflects the traffic cost of the aircraft from any position to the destination. And then selecting an optimal path with the minimum cost according to the traffic value graph, wherein the path is a flight route control point sequence formed by grid coordinates.
3) Path smoothing system
And after the route smoothing system obtains the route processed by the route planning system, a route smoothing algorithm smoothes the route control point sequence according to the physical characteristics and the dynamic characteristics of the aircraft, and finally, the smoothed result is sent to the airborne platform.
As shown in fig. 2, the specific steps are as follows:
firstly, an aircraft deploys an airborne platform, four-dimensional track data are collected, flight starting point, end point and obstacle information are processed into a dynamic point track sequence, and the dynamic point track sequence is sent to a route planning module;
as shown in fig. 1, the airborne platform S11 is responsible for collecting four-dimensional track data (longitude, latitude, altitude, time) and airborne radar data, processing the obstacle data into a dynamic track sequence, and issuing the start point, the end point and the dynamic obstacle data of the flight to the path planning system S12.
And the airborne platform S11 is also responsible for finally receiving the planned track data and controlling the collision-free flight of the aircraft according to the track data.
And step two, constructing a four-dimensional space-time airspace grid by the route planning module according to the coordinate relation of different three-dimensional spaces on each time slice.
When the environment changes dynamically, the three-dimensional grid map cannot completely describe the movement condition of the obstacle in the environment. It is therefore necessary to expand the three-dimensional grid map into a four-dimensional grid map along the time axis. After obtaining the starting point, the ending point and the dynamic obstacle information of the airborne platform S11, the path planning system S12 firstly needs to construct a four-dimensional space-time airspace grid according to the coordinate relation of different three-dimensional spaces on each time slice.
The specific construction process is as follows:
and generating a three-dimensional target airspace by taking a starting point and a terminal point of the aircraft as boundaries, dividing the three-dimensional target airspace into a plurality of four-dimensional grids, setting the length, the width and the height of each four-dimensional grid unit according to the standard interval distance of the aircraft, and corresponding the flight paths of the aircraft and the obstacles to the set four-dimensional grids according to time slices.
Thirdly, based on the airspace grid, calculating a four-dimensional space-time passing value graph according to reinforcement learning iteration; and selecting a conflict-free path with the minimum cost from the traffic value graph, and conveying the conflict-free path to the path smoothing module.
According to the four-dimensional space-time airspace grid map, the aircraft traverses from each position to reach the destination, and the passing cost corresponding to each position is calculated; all the passing costs form a four-dimensional space-time passing value graph;
after the space grid of the four-dimensional space-time is constructed, the traffic value map of the four-dimensional space-time can be calculated according to reinforcement learning.
The reinforcement learning defines the strategy pi as: mapping of the state s to the probability of any feasible action a; the choice of strategy pi is varied.
The value v(s) of the state s is defined as: the expectation of rewards earned by the aircraft for reaching the target location along a path selected using strategy pi, starting from state s; wherein the reward r(s) is feedback that the agent gives in the context of selecting action a in state s to encourage or inhibit that action;
in a four-dimensional spatiotemporal traffic value graph, the states s= (t, i, j, k) are collectively described by time t and grid position (i, j, k). The optimal path planned by the traffic value graph is also described by a state sequence consisting of time-position quaternions:
(t 0 ,i 0 ,j 0 ,k 0 )→(t 1 ,i 1 ,j 1 ,k 1 )→…→(t N ,i goal ,j goal ,k goal )。
since time is flowing in the forward direction, in one path sequence, two adjacent path points s t Sum s t+1 In addition to being adjacent in position, the time also increases in sequence: (t) 0 ,i 0 ,j 0 ,k 0 )→(t 0 +1,i 1 ,j 1 ,k 1 )→…→(t 0 +N,i goal ,j goal ,k goal )
Thus in four dimensions time-space, action a Δi,Δj,Δk The state transitions caused are defined as:
s'＝trans(s,a Δi,Δj,Δk )＝<s.t+1,s.i+Δi,s.j+Δj,s.k+Δk>
i.e. the state moves in the four-dimensional space in the time increasing direction towards adjacent positions.
The definition of the return function r(s) is:
wherein S is obstacle Set of occupied grid locations for other elements (obstacles) in the environment other than the current agent
{<i,j,k> obstacle1 ,<i,j,k> obstacle2 ,..}, due to the grid mapIs time-varying, and thus the set of states S occupied by the obstacle obstacle And also over time. Therefore, in calculating the return of the state s, the set of obstacles in the grid map at the time s.t where s is located needs to be selected
First, a four-dimensional rewards graph is calculated from a four-dimensional grid map, the values in the rewards graph representing rewards values given when time t is at grid position (i, j, k). And then carrying out value iteration on the traffic value graph based on the return graph.
In the process of value iteration, for any state s= (t, i, j, k), the next adjacent state { t+1, i+Δi, j+Δj, k+Δk } is selected, where Δi, Δj, Δk e { -1,0,1}. The traffic value of the state (t, i, j, k) is updated afterwards.
The value of all positions except the target position at all times needs to be set to- + -infinity in the initialization stage of the iterative algorithm, the value of the target position at all times is set to 0. Let N be the maximum input frame number of the value iterative algorithm.
When calculating, firstly, calculating a row value diagram at the last moment by using a grid map of the last frame in a static map mode by using a three-dimensional space value iterative algorithm. And after the traffic value graph at the last moment converges, calculating the traffic value graph at other moments by using a four-dimensional space-time value iteration method. Since the last frame of traffic value graph is calculated and converged by using a three-dimensional space value iterative algorithm, traffic values of all positions in the traffic value graph at other moments based on the calculation are updated finally.
When the value of each state in the traffic value graph is the optimal value, a path optimal path can be selected according to the traffic value graph, and finally the path is issued to the route smoothing system S13.
Based on a defined state function r(s) and a return function v(s), calculating the return of each position by using the return function according to input grid track data, a four-dimensional space-time airspace grid map and flight starting and ending point data, storing the return into a return report map, and initializing a value map; and then calculating the return at each moment by using a value iterative algorithm based on the return graph, and calculating a three-dimensional space-time traffic value graph converged at the moment T through multiple iterations. According to the three-dimensional space-time traffic value graph at the time T, a round of iteration is sequentially carried out to obtain four-dimensional space-time traffic value graphs at the time T-1, T-2 and … and at the time 0, and traffic values of all positions are updated finally.
In the method, a space-time value iterative algorithm is realized in a mode of a nerve convolution neural network, and a cost function is updated according to the following formula:
and after the traffic value graph is obtained, calculating the optimal flight route according to the traffic value graph.
And step four, after the route smoothing module obtains a conflict-free route with the minimum cost, a route smoothing algorithm smoothes a route control point sequence according to physical characteristics and dynamic characteristics of the aircraft, and finally, the smoothed result is sent to an airborne platform, and the airborne platform controls the flight of the aircraft according to the route.
The calculation of the optimal flight route is divided into two parts of route control point selection and route smoothing:
the route smoothing system S13 firstly obtains an optimal route processed by the route planning system S12, the route is a flight route control point sequence composed of grid coordinates, and then a route smoothing algorithm smoothes the route control point sequence according to the physical characteristics and the dynamic characteristics of the aircraft; the data smoothing adopts a linear interpolation method.
When selecting the optimal path according to the traffic value graph, the current position is represented by the state s 0 ＝(t 0 ,i 0 ,j 0 ,k 0 ) Starting, sequentially selecting the subsequent states as route control points according to the following formula until the target state is reached:
where s' ∈trans (s, a) is the neighbor state of state s.
Although the route smoothing algorithm can smooth the route control point sequence according to the physical characteristics and the dynamic characteristics of the agent, when the control point has an excessively sharp corner, the smoothed flight route may deviate from the control point far, so that the agent cannot run strictly according to the flight route planned by the route control point, and the traffic safety is affected; and sharp corners in the route also severely impact the riding experience of the agent.
Therefore, the corner transformation angle and the traffic value can be used as the basis for selecting the route control point. The route control point selection formula is rewritten as:
where θ is the orientation at state s. Initial state s 0 Is determined based on the current orientation of the agent. The orientation of the adjacent state s 'is calculated from the orientation of s and the relative position of s and s'.
The orientation when calculating state s' is therefore calculated using the following formula:
function f α The effect of the change in the angle of rotation caused by the transition from state s to state s' on the traffic value is calculated. Function f α It should be less sensitive to changes in turn as the agent's turn performance is better and more sensitive to changes in turn as the occupant's value preference is more conservative, depending on the agent model and occupant flight preference.
After the route smoothing algorithm is processed, the route smoothing system S13 can send the smoothed track data to the airborne platform S11, and the airborne platform can further control the flight of the aircraft after obtaining the planned track data.
The core modules in the proposal of the application are a path planning system and a path smoothing system:
1) Path planning system
The purpose of the path planning of the aircraft is to obtain an optimal collision-free path of travel. When the environment is changed drastically, the number of obstacles in the airspace is relatively large and the obstacles change dynamically with time, it is very difficult to plan a flight path with short starting point-to-ending point distance, little time consumption and no collision. Since the three-dimensional grid map cannot fully describe the movement of obstacles in the environment. Therefore, the three-dimensional grid map is expanded into a four-dimensional space-time airspace grid along a time axis, a four-dimensional space-time traffic value map is calculated on the basis of four-dimensional space-time data, information such as the movement situation of each obstacle in the current environment is processed into a value map form, and an optimal collision-free route is planned from a global angle according to the value map.
2) Path smoothing system
The path smoothing has great importance to the route planning, and when the route has too sharp corners, the riding experience of personnel in the aircraft can be seriously influenced, and even the air traffic safety is influenced. After the sequence after the path planning is obtained, the system can smooth the route control point sequence according to the physical characteristics and the dynamic characteristics of the aircraft, so that the finally planned track is ensured to be stable and reliable.
Claims (3)
1. A conflict-free route planning method based on airspace grids is characterized by comprising the following specific steps:
firstly, an aircraft deploys an airborne platform, four-dimensional track data are collected, flight starting point, end point and obstacle information are processed into a dynamic point track sequence, and the dynamic point track sequence is sent to a route planning module;
step two, constructing a four-dimensional space-time airspace grid by the route planning module according to the coordinate relation of different three-dimensional spaces on each time slice;
thirdly, based on the airspace grid, calculating a four-dimensional space-time passing value graph according to reinforcement learning iteration; selecting a conflict-free path with the minimum cost from the traffic value graph, and conveying the conflict-free path to a path smoothing module;
the specific process is as follows:
step 301, a cost function v(s) and a return function r(s) of a defined state s;
the cost function v(s) is defined as: the expectation of rewards earned by the aircraft for reaching the target location along a path selected using strategy pi, starting from state s; policy pi is defined as the mapping of state s to the probability of any feasible action a;
the definition of the return function r(s) is:
wherein S is obstacle An occupied set of grid locations for other elements in the environment other than the current agent; s is S goal Grid location for the target;
step 302, traversing the paths of the aircraft from each position to the destination according to the four-dimensional space-time airspace grid map, calculating the return, namely the passing cost, of each position by using a return function, and storing the return into a return map;
for any state s= (t, i, j, k), the values in the reward graph represent the reward values given when time t is at grid position (i, j, k);
step 303, calculating returns at all moments based on the return graphs by using a value iterative algorithm, and calculating a four-dimensional space-time passing value graph converged at the moment T through multiple iterations;
firstly, in the initialization stage of an iterative algorithm, setting the value of all positions except the target position at all times to be- ≡, and setting the value of the target position at all times to be 0; n is the maximum input frame number of the value iterative algorithm;
then, during calculation, a three-dimensional space value iterative algorithm is used for calculating a traffic value graph at the moment T by using a grid map of the last frame in a static map mode;
after the passing value map at the last moment T converges, a four-dimensional space-time value iterative method is used for sequentially calculating to obtain the passing value maps at the moments T-1, T-2, … and 0, and the passing values of all positions are updated finally;
the space-time cost function is updated according to the following formula:
step 304, when the value of each state in the traffic value graph is the optimal value, the selected optimal flight route path is the optimal flight route path;
the path is a flight route control point sequence composed of grid coordinates;
and step four, after the route smoothing module obtains a conflict-free route with the minimum cost, a route smoothing algorithm smoothes a route control point sequence according to physical characteristics and dynamic characteristics of the aircraft, and finally, the smoothed result is sent to an airborne platform, and the airborne platform controls the flight of the aircraft according to the route.
2. The collision-free route planning method based on airspace grids according to claim 1, wherein the specific construction process in the step two is as follows:
and generating a three-dimensional target airspace by taking a starting point and an ending point of the aircraft as boundaries, dividing the three-dimensional target airspace into a plurality of four-dimensional grids, setting the length, the width and the height of each four-dimensional grid unit according to the standard interval distance of the aircraft, and corresponding the positions of the flight path and the obstacle of the aircraft to the set four-dimensional grids according to time slices.
3. A collision-free route planning method based on airspace network as claimed in claim 1, in which in said step four, the data smoothing adopts a linear interpolation method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310477933.9A CN116518974A (en) | 2023-04-28 | 2023-04-28 | Conflict-free route planning method based on airspace grids |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310477933.9A CN116518974A (en) | 2023-04-28 | 2023-04-28 | Conflict-free route planning method based on airspace grids |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116518974A true CN116518974A (en) | 2023-08-01 |
Family
ID=87395432
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310477933.9A Pending CN116518974A (en) | 2023-04-28 | 2023-04-28 | Conflict-free route planning method based on airspace grids |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116518974A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116978261A (en) * | 2023-09-25 | 2023-10-31 | 粤港澳大湾区数字经济研究院(福田) | Space-time resource and space-time process management system and flight scheduling method |
| CN117519278A (en) * | 2023-12-04 | 2024-02-06 | 上海市建筑科学研究院有限公司 | Unmanned aerial vehicle obstacle avoidance method for bridge inspection |
-
2023
- 2023-04-28 CN CN202310477933.9A patent/CN116518974A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116978261A (en) * | 2023-09-25 | 2023-10-31 | 粤港澳大湾区数字经济研究院(福田) | Space-time resource and space-time process management system and flight scheduling method |
| CN116978261B (en) * | 2023-09-25 | 2024-04-09 | 粤港澳大湾区数字经济研究院(福田) | Space-time resource and space-time process management system and flight scheduling method |
| CN117519278A (en) * | 2023-12-04 | 2024-02-06 | 上海市建筑科学研究院有限公司 | Unmanned aerial vehicle obstacle avoidance method for bridge inspection |
| CN117519278B (en) * | 2023-12-04 | 2024-04-30 | 上海市建筑科学研究院有限公司 | Unmanned aerial vehicle obstacle avoidance method for bridge inspection |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110673637B (en) | Unmanned aerial vehicle pseudo path planning method based on deep reinforcement learning | |
| CN111780777B (en) | Unmanned vehicle route planning method based on improved A-star algorithm and deep reinforcement learning | |
| CN116518974A (en) | Conflict-free route planning method based on airspace grids | |
| CN110470301B (en) | Unmanned aerial vehicle path planning method under multi-dynamic task target point | |
| US20210370980A1 (en) | Autonomous vehicle planning | |
| KR102306939B1 (en) | Method and device for short-term path planning of autonomous driving through information fusion by using v2x communication and image processing | |
| CN109521794A (en) | A kind of multiple no-manned plane routeing and dynamic obstacle avoidance method | |
| Ma et al. | Improved ant colony algorithm for global optimal trajectory planning of UAV under complex environment. | |
| CN112034887A (en) | Optimal path training method for unmanned aerial vehicle to avoid cylindrical barrier to reach target point | |
| CN114839968B (en) | Path planning method for unmanned surface vehicle | |
| CN112327939B (en) | Collaborative path planning method for high-rise fire-fighting multiple unmanned aerial vehicles in city block environment | |
| CN114610066A (en) | Method for generating formation flight tracks of distributed cluster unmanned aerial vehicles in complex unknown environment | |
| CN116795108B (en) | Intelligent unmanned vehicle distribution method based on multi-source sensing signals | |
| CN115047910A (en) | Unmanned aerial vehicle formation cruise control method based on wild goose-shaped array | |
| CN110793522A (en) | Flight path planning method based on ant colony algorithm | |
| CN115494866A (en) | Multi-unmanned aerial vehicle global and local path intelligent planning method and system | |
| CN116795138A (en) | Data acquisition-oriented intelligent track planning method for multiple unmanned aerial vehicles | |
| CN117029827A (en) | Unmanned aerial vehicle track planning method and device under wind and rain conditions | |
| CN116385909A (en) | Unmanned aerial vehicle target tracking method based on deep reinforcement learning | |
| CN116718190A (en) | Mobile robot path planning method in long-distance dense crowd scene | |
| Ma et al. | Adaptive deployment of UAV-aided networks based on hybrid deep reinforcement learning | |
| CN112099521B (en) | Unmanned aerial vehicle path planning method and device | |
| CN115237157A (en) | Road network constraint-based air-to-ground unmanned cluster multi-task point path planning method | |
| CN114779820A (en) | Multi-destination unmanned aerial vehicle real-time flight path planning method with intelligent decision | |
| Saska et al. | Control of ad-hoc formations for autonomous airport snow shoveling |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |