CN115328205B - Flying vehicle takeoff and landing decision planning method based on three-dimensional target detection - Google Patents

Abstract

The invention discloses a flying automobile takeoff and landing decision planning method based on three-dimensional target detection, which is realized based on a camera and a laser radar which are deployed on a flying automobile, and comprises the following steps: establishing a dynamic three-dimensional map according to the real-time acquired image and the three-dimensional point cloud data; when the flying automobile is in a flying state, respectively establishing a repulsive field and a gravitational field to form a real-time changing resultant force field according to the dynamic three-dimensional map, driving the flying automobile to dynamically land, controlling the flying automobile to land when a safe distance is met, and finishing decision planning; otherwise, controlling the flying automobile to ascend, and reestablishing the gravitational field; and when the flying automobile is in a running state, respectively establishing a repulsive force field and a gravitational field to form a real-time changing resultant force field according to the dynamic three-dimensional map, driving the flying automobile to take off, and finishing the decision planning.

Description

A flying car takeoff and landing decision planning method based on three-dimensional target detection

Technical Field

The present invention belongs to the technical field of autonomous driving, and in particular relates to a take-off and landing decision-making planning method for a flying car based on three-dimensional target detection.

Background Art

Henry Ford, the founder of Ford Motor Company, once predicted that a combination of airplanes and cars would soon be available. As urban traffic congestion becomes increasingly serious, urban air traffic represented by flying cars is a new solution. Many manufacturers have set around 2025 as an important time node for the commercialization of flying cars.

Two questions arise:

1) How to make a flying car with initial velocity merge safely into the traffic flow on a busy road;

2) How to make a flying car with initial velocity take off safely from the busy ground.

It is much more difficult to merge into traffic from the air than to fly from traffic into the air. The take-off field is a simplified form of the landing field.

In order to solve the above problems, two issues need to be solved: environmental perception and path planning:

Ordinary cameras can only distinguish the target category when detecting targets, and the position perception error of the target in three-dimensional space is large. LiDAR has a natural distance advantage in three-dimensional target detection and can directly obtain the position information of each point, but it lacks rich visual information of images.

Different types of sensors have their own advantages and disadvantages. A single sensor cannot achieve accurate and efficient detection. To this end, multiple sensors with complementary characteristics are fused to enhance perception capabilities. The three-dimensional target detection method that fuses images and point clouds can make up for the lack of depth in images as well as the lack of visual information in point clouds.

The current path planning algorithms are divided into two categories, the global path planning algorithms represented by the A* algorithm, and the local obstacle avoidance algorithms represented by the artificial potential field method. Both algorithms have their own advantages and disadvantages. The A* algorithm can obtain the global optimal solution to prevent the drone from falling into the local optimal solution, but the A* algorithm needs to know the information of the entire map in advance and the algorithm will take longer to solve as the map increases; while the artificial potential field method can quickly respond to the obstacle position information, the method is highly reliable, does not rely on the prior information of the environment and the shape of the obstacle, and is not affected by the shape of the obstacle, but it will fall into the local optimum; specifically, the basic principle of the artificial potential field method is to generate two virtual force fields during the flight: the gravitational field and the repulsive field. Then, under the joint action of the two force fields, different forces are generated according to the different models of each force field, and the flying car takes off and lands safely under the action of these forces.

Summary of the invention

The purpose of the present invention is to overcome the defects of the prior art and propose a flying car take-off and landing decision planning method based on three-dimensional target detection.

In order to achieve the above-mentioned object, the present invention proposes a flying car takeoff and landing decision planning method based on three-dimensional target detection, which is implemented based on a camera and a laser radar deployed on the flying car. The method comprises:

Step 1) Perform environmental perception on a designated road based on real-time collected images and three-dimensional point cloud data to create a dynamic three-dimensional map;

Step 2) When the flying car to be planned and decided is in the flying state, go to step 3); when the flying car to be planned and decided is in the driving state, go to step 5);

Step 3) Based on the dynamic three-dimensional map, combined with the flying car data and the situation of cars driving on the one-way road, a repulsive field is established, and a point behind the driving car or a point on the ground is selected to establish a gravitational field;

Step 4) The repulsive field and the gravitational field form a real-time changing resultant force field to drive the flying car to perform dynamic landing, and determine whether the distance between the flying car and the target position meets the safety distance. If it is determined to be yes, the flying car is controlled to land and the process goes to step 7); if it is determined to be no, the flying car is controlled to ascend and the process goes to step 3) to re-establish the gravitational field;

Step 5) Based on the dynamic three-dimensional map, combined with the flying car data and the situation of cars traveling on the one-way road, a repulsive field is established, and a safe position in the air is selected to establish a gravitational field;

Step 6) The repulsive field and the gravitational field form a real-time changing resultant force field to drive the flying car to take off, and then go to step 7);

Step 7) Decision planning is completed.

As an improvement of the above method, the step 1) comprises:

Step 1-1) The collected image is processed by a convolutional neural network and image pyramid feature extraction is performed to obtain an image feature map of the same size as the initial image;

Step 1-2) The acquired three-dimensional point cloud and image feature map are jointly calibrated to obtain a point cloud within the image range and obtain corresponding image features;

Step 1-3) voxel-gridding the fused point cloud image data according to the distribution of the point cloud to obtain voxelized data;

Step 1-4) Filter the voxelized data, remove the empty grids, and arrange the length, width, and height in order into one dimension to obtain the processed voxelized data.

Step 1-5) Input the processed voxelized data into the data encoding network to obtain a feature map;

Step 1-6) The feature map is passed through a single-stage target detection network to obtain the coordinates and length, width and height information of the three-dimensional target on the specified road;

Step 1-7) Create a dynamic three-dimensional map based on the coordinates and length, width and height information of each detected three-dimensional target.

As an improvement of the above method, the data encoding network of step 1-5) includes: a fully connected layer and VoxelNet; the processing of the data encoding network includes:

The processed voxelized data includes several non-empty grids, each grid includes several points, a point is extracted from each grid to represent the grid, and a grid is selected in a height direction to represent the height, and a feature map size of L length and W width is obtained. C is the feature number of the feature map, and the features are expanded by increasing the dimension.

As an improvement of the above method, the step 3) comprises:

Step 3-1) extracting the data of the flying car to be planned and analyzing and processing it to establish a repulsive field;

The repulsive field U _ri (X) and repulsive force F _r (X) acting on the flying car satisfy the following equation:

Wherein, _kr represents the positive proportional coefficient, X represents the position of the flying car to be planned, _Xi represents the position of obstacle i, _ηi (X, _Xi ) represents the distance between X and _Xi , _ηi (X, _Xi ) represents the distance between X and _Xi , and η0 _,i represents the maximum distance of the repulsive field of the i-th obstacle;

Where N represents the total number of obstacles, and _Fri (X) represents the repulsive force of obstacle i, satisfying the following formula:

表示η_i梯度；in,

represents η _i gradient;

Step 3-2) Establish a gravitational field based on the vehicle conditions in front and behind the flying car to be planned;

The gravitational field U _a (X) and gravitational force F _a (X) acting on the flying car satisfy the following equation:

表示η梯度。Among them, ρ represents the gravitational proportional coefficient, X _g represents the target position, η(X,X _g ) represents the distance between the flying car and the target position,

represents the gradient of η.

As an improvement of the above method, the step 4) determining whether the distance between the flying car and the target position meets the safety distance comprises:

When the distance between the front vehicle and the rear vehicle is greater than or equal to the safety distance, the safety distance is met, otherwise the safety distance is not met.

As an improvement of the above method, the step 5) comprises:

Step 5-1) extracting the data of cars traveling on a one-way road and the flight data of the flying car to be planned and decided, and establishing a repulsive field after analysis and processing;

The repulsive field U _ri (X) and repulsive force F _r (X) acting on the flying car satisfy the following equation:

Wherein, _kr represents the positive proportional coefficient, X represents the position of the flying car to be planned, _Xi represents the position of obstacle i, _ηi (X, _Xi ) represents the distance between X and _Xi , _ηi (X, _Xi ) represents the distance between X and _Xi , and η0 _,i represents the maximum distance of the repulsive field of the i-th obstacle;

Step 5-2) Select a safe position in the air to establish a gravitational field;

The gravitational field U _a (X) and gravitational force F _a (X) acting on the flying car satisfy the following equation:

表示η梯度。Among them, ρ represents the gravitational proportional coefficient, X _g represents the target position, η(X,X _g ) represents the distance between the flying car and the target position,

represents the gradient of η.

A flying car takeoff and landing decision-making planning system based on three-dimensional target detection is implemented based on a camera and a laser radar deployed on the flying car, characterized in that the system includes: a dynamic three-dimensional map construction module, a state judgment module, a flight repulsion gravitational field establishment module, a landing control module, a driving repulsion gravitational field establishment module and a takeoff control module;

The dynamic three-dimensional map building module is used to perform environmental perception on the designated road based on the real-time collected images and three-dimensional point cloud data to build a dynamic three-dimensional map;

The state judgment module is used to switch to the flight repulsive gravitational field establishment module when the flying car to be decided and planned is in the flying state; and to switch to the driving repulsive gravitational field establishment module when the flying car to be decided and planned is in the driving state;

The flight repulsive gravitational field establishment module is used to establish a repulsive field based on the dynamic three-dimensional map, combined with the flying car data and the situation of cars traveling on a one-way road, and select a point behind the traveling car or a point on the ground to establish the gravitational field;

The landing control module is used to form a real-time changing resultant force field by the repulsive field and the gravitational field, drive the flying car to perform dynamic landing, and judge whether the distance between the flying car and the target position meets the safety distance. If it is judged to be yes, the flying car is controlled to land and the decision-making planning is ended; if it is judged to be no, the flying car is controlled to ascend and the gravitational field is reestablished in the flight repulsive gravitational field establishment module;

The driving repulsive gravitational field establishment module is used to establish a repulsive field based on the dynamic three-dimensional map, combined with the flying car data and the situation of cars driving on a one-way road, and select a safe position in the air to establish a gravitational field;

The take-off control module is used to form a real-time changing resultant force field from the repulsive field and the gravitational field to drive the flying car to take off, and the decision-making planning is completed.

Compared with the prior art, the advantages of the present invention are:

1. Different types of sensors have their own advantages and disadvantages. A single sensor cannot achieve accurate and efficient detection. The present invention fuses multiple sensors with complementary characteristics to enhance perception capabilities. The three-dimensional target detection method of image and point cloud fusion can make up for the lack of depth in the image and the lack of visual information in the point cloud.

2. A dynamic 3D map is established based on the vehicle coordinates and length, width and height information provided by the 3D target detection method of image and point cloud fusion. The coordinates and length, width and height information of the vehicle in space are reflected in the dynamic 3D map in real time;

3. Based on the dynamic three-dimensional map, the gravitational field and repulsive field are established, and the flying car lands and takes off under the action of the real-time changing potential field.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG1 is a landing control decision flow chart of a takeoff and landing decision planning method for a flying car based on three-dimensional target detection according to the present invention;

FIG2 is a takeoff control decision flow chart of the takeoff and landing decision planning method of a flying car based on three-dimensional target detection of the present invention.

DETAILED DESCRIPTION

To solve the above problems, the present invention provides a flying car takeoff and landing decision planning method based on three-dimensional target detection. Using the dynamic three-dimensional map established by the data of three-dimensional target detection, the flying car can safely merge into the traffic flow from the air with an initial velocity, or take off from the traffic flow with an initial velocity under the action of the dynamic potential field.

The technical solution of the present invention is described in detail below with reference to the accompanying drawings and embodiments.

Example 1

Embodiment 1 of the present invention proposes a flying car takeoff and landing decision planning method based on three-dimensional target detection. After obtaining road information through cameras and laser radars, the flying car can safely merge from the air into the motor vehicle lane in the city.

The method for detecting dynamic landing sites based on three-dimensional targets mainly includes the following steps:

As shown in Figure 1, the specific implementation steps of the dynamic landing field are as follows:

Step 1: The image captured by the camera is processed by a convolutional neural network and extracted through an image pyramid to obtain an image feature map of the same size as the initial image.

The image (h, w, 3) means that the length is w pixels, the width is h pixels, and each pixel has 3 RGB channels; the image feature map (h, w, n _c ) means that the length is w pixels, the width is h pixels, and each pixel has n _c channels;

Step 2: The coordinate transformation matrix of the 3D point cloud acquired by the lidar is jointly calibrated and transformed to the image coordinate system for projection. Each point cloud within the image size range obtains the features of the corresponding pixel points, and finally obtains the point cloud within the image range and obtains the corresponding image features.

The 3D point cloud (n, 4) means that n point clouds are acquired in one frame, and each point cloud has four channels (x, y, z, r), where (x, y, z) is the 3D coordinate of the point cloud and r is the reflectivity of the point cloud; the image feature ( _nr , 4+ _nc ) means _nr points within the image range, and 4+ _nc means the original 4 (x, y, z, r) features of the point cloud are spliced with the image feature _nc of the corresponding point

Step 3: voxel-grid the fused point cloud image data according to the distribution of the point cloud to obtain voxelized data;

进行升序排列，取前N个点。The voxelized data (L, W, H, N, 4+n _c ) represents L grids on the long side, W grids on the wide side, H grids on the height, N point cloud image data in each grid, 4+n _c feature channels for each data, and N has a maximum limit. If there are more than N points in the grid, then according to the xyz coordinates of the point, according to the distance from the origin

Arrange in ascending order and take the first N points.

Step 4: Filter the voxelized data, remove the empty grids, and arrange the length, width, and height in order into one dimension to obtain the voxelized data.

The filtered voxelized data (K, N, 4 + n _c represents K non-empty grids, N data points per grid, and 4 + n _c feature channels per data.

Step 5: Input the processed voxelized data into the data encoding network to obtain the feature map.

The data encoding network includes a fully connected layer and VoxelNet. The data encoding network extracts a point in each grid to represent the grid, and selects a grid in a height direction to represent the height, that is, a feature map size of L length and W width is obtained. C is the number of feature maps, and features are expanded by increasing the dimension.

Step 6: Pass the feature map through a single-stage target detection network, and the final output layer directly outputs the three-dimensional target (x, y, z, ry, l, w, h, s).

The (x, y, z, ry, l, w, h, s) of a three-dimensional target represents the (x, y, z) coordinates of the target, the angle ry between its orientation and the x-axis of the origin coordinate system, the length, width and height of the target (l, w, h), and the confidence score s.

Step 7: Create a dynamic 3D map based on the detected 3D vehicle target (x, y, z, ry, l, w, h, s).

Step 8 extracts the data of the car traveling on the one-way road, and establishes a repulsive potential field after analyzing and processing the data of the car traveling on the one-way road.

The repulsive field and repulsive force acting on the flying car:

表示梯度Where k _r >0, represents the positive proportional coefficient, X represents the position of the flying car, _Xi represents the position of obstacle i, η _0,i represents the maximum distance of the repulsive field of the i-th obstacle,

Represents the gradient

Step 9: The decision system selects a point behind the moving car or a point on the ground to establish a gravitational field.

Gravitational field and gravitational force acting on a flying car:

表示梯度Among them, ρ>0, represents the gravity proportional coefficient, X represents the position of the flying car, _Xg represents the position of the target, and η(X, _Xg ) represents the distance between the flying car and the target.

Represents the gradient

Step 10: The gravitational field and the repulsive field form a resultant force field that changes in real time, and the flying car performs dynamic landing under the action of the resultant force field that changes in real time.

Step 11: If η(X ₁ , X ₂ ) < safe distance, the flying car rises to a safe position under the action of the gravitational field and the repulsive field, and returns to step 8 to re-establish the gravitational field. If η(X ₁ , X ₂ ) ≥ safe distance, the flying car lands safely under the action of the gravitational field and the repulsive field.

_X1 represents the position of the front vehicle, _X2 represents the position of the rear vehicle, and η( _X1 , _X2 ) represents the distance between the front vehicle and the rear vehicle.

As shown in Figure 2, the specific implementation steps of the dynamic take-off field are as follows:

Steps 1-7 are consistent with the landing site;

Step 8 extracts the data of cars traveling on the one-way road and cars flying in the air, and establishes a repulsive potential field after analyzing and processing the data of cars traveling on the one-way road and cars flying in the air.

The repulsive field and repulsive force acting on the flying car:

表示梯度，需要说明，对于起飞和降落斥力场公式相同，但是k_r取值不同。Where k _r >0, represents the positive proportional coefficient, X represents the position of the flying car, _Xi represents the position of obstacle i, η _0,i represents the maximum distance of the repulsive field of the i-th obstacle,

Represents the gradient. It should be noted that the formula for the repulsive field is the same for takeoff and landing, but the values of k _r are different.

Step 9: The decision system selects a safe position in the air to establish a gravitational field;

Gravitational field and gravitational force acting on a flying car:

表示梯度Among them, ρ>0, represents the gravity proportional coefficient, X represents the position of the flying car, _Xg represents the position of the target, and η(X, _Xg ) represents the distance between the flying car and the target.

Represents the gradient

Step 10: The gravitational field and the repulsive field form a resultant force field that changes in real time, and the flying car takes off dynamically under the action of the resultant force field that changes in real time.

Example 2

Embodiment 2 of the present invention proposes a flying car takeoff and landing decision-making planning system based on three-dimensional target detection, which is implemented based on a camera and a laser radar deployed on the flying car. The system includes: a dynamic three-dimensional map construction module, a state judgment module, a flight repulsion field establishment module, a landing control module, a driving repulsion field establishment module and a takeoff control module; the method of embodiment 1 is used to implement it.

The dynamic three-dimensional map building module is used to perform environmental perception on the designated road based on the real-time collected images and three-dimensional point cloud data to build a dynamic three-dimensional map;

The state judgment module is used to switch to the flight repulsive gravitational field establishment module when the flying car to be decided and planned is in the flying state; and to switch to the driving repulsive gravitational field establishment module when the flying car to be decided and planned is in the driving state;

The flight repulsive gravitational field establishment module is used to establish a repulsive field based on the dynamic three-dimensional map, combined with the flying car data and the situation of cars traveling on a one-way road, and select a point behind the traveling car or a point on the ground to establish the gravitational field;

The landing control module is used to form a real-time changing resultant force field by the repulsive field and the gravitational field, drive the flying car to perform dynamic landing, and judge whether the distance between the flying car and the target position meets the safety distance. If it is judged to be yes, the flying car is controlled to land and the decision-making planning is ended; if it is judged to be no, the flying car is controlled to ascend and the gravitational field is reestablished in the flight repulsive gravitational field establishment module;

The driving repulsive gravitational field establishment module is used to establish a repulsive field based on the dynamic three-dimensional map, combined with the flying car data and the situation of cars driving on a one-way road, and select a safe position in the air to establish a gravitational field;

The take-off control module is used to form a real-time changing resultant force field by the repulsive field and the gravitational field to drive the flying car to take off, and the decision-making planning is completed.

Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention is described in detail with reference to the embodiments, it should be understood by those skilled in the art that any modification or equivalent replacement of the technical solutions of the present invention does not depart from the spirit and scope of the technical solutions of the present invention and should be included in the scope of the claims of the present invention.

Claims (7)

1. A flying car takeoff and landing decision-making planning method based on three-dimensional target detection is realized based on a camera and a laser radar which are deployed on a flying car, and comprises the following steps:

step 1) carrying out environmental perception on a specified road according to a real-time acquired image and three-dimensional point cloud data, and establishing a dynamic three-dimensional map;

step 2) when the hovercar to be planned by decision-making is in a flight state, turning to step 3); when the flying automobile to be planned by the decision-making is in a running state, turning to the step 5);

step 3) establishing a repulsive field according to the dynamic three-dimensional map by combining the data of the flying automobile and the condition of the automobile running on a one-way road, and selecting a certain point behind the running automobile or a certain point on the ground to establish a gravitational field;

step 4), forming a real-time changing resultant force field by the repulsive force field and the gravitational field, driving the hovercar to dynamically land, judging whether the distance between the hovercar and the target position meets the safety distance, if so, controlling the hovercar to land, and turning to the step 7); if not, controlling the aerocar to ascend, and turning to the step 3) to reestablish the gravitational field;

step 5) establishing a repulsive field according to the dynamic three-dimensional map by combining the data of the aerocar and the situation of the car running on a one-way road, and selecting an air safety position to establish a gravitational field;

step 6), forming a real-time changing resultant force field by the repulsive force field and the gravitational field, driving the aerocar to take off, and turning to the step 7);

and 7) finishing decision planning.

2. The flying vehicle takeoff and landing decision planning method based on three-dimensional target detection as claimed in claim 1, wherein the step 1) comprises:

step 1-1), processing the acquired image by a convolutional neural network, and extracting image pyramid characteristics to obtain an image characteristic diagram with the same size as the initial image;

step 1-2) carrying out combined calibration on the obtained three-dimensional point cloud and the image characteristic graph to obtain a point cloud in an image range and obtain corresponding image characteristics;

step 1-3) carrying out voxel meshing on the fused point cloud image data according to the distribution of the point cloud to obtain voxel data;

step 1-4) screening the voxelized data, removing empty grids, arranging the length, the width and the height into a dimension according to a sequence to obtain the processed voxelized data,

step 1-5) inputting the processed voxelized data into a data coding network to obtain a characteristic diagram;

step 1-6) passing the feature map through a single-stage target detection network to obtain the coordinate, length, width and height information of the three-dimensional target of the specified road;

and 1-7) establishing a dynamic three-dimensional map according to the detected coordinate, length, width and height information of each three-dimensional target.

3. The flying vehicle takeoff and landing decision-making planning method based on three-dimensional target detection as claimed in claim 2, wherein the data coding network of the step 1-5) comprises: a full connection layer and a VoxelNet; the processing of the data encoding network includes:

the processed voxelized data comprises a plurality of non-empty grids, each grid comprises a plurality of points, one point is extracted from each grid to represent the grid, one grid is selected from one height direction to represent the height, the size of a feature map with the length L and the width W is obtained, C is a feature number of the feature map, and features are expanded through ascending dimensions.

4. The flying vehicle takeoff and landing decision-making planning method based on three-dimensional target detection as claimed in claim 1, wherein the step 3) comprises:

step 3-1) extracting data of the aerocar to be planned by decision, and establishing a repulsive force field after analysis and processing;

repulsive force field U acting on aerocar _ri (X) and repulsive force F _r (X) satisfies the following formula:

wherein k is _r Expressing a direct proportionality coefficient, X expressing the position of the hovercar to be planned by decision, X _i Indicates the position, η, of the obstacle i _i (X,X _i ) Denotes X and X _i A distance therebetween, η _i (X,X _i ) Represents X and X _i A distance therebetween, η _0，i Represents the maximum distance acted by the repulsive force field of the ith obstacle;

wherein N represents the total number of obstacles, F _ri (X) represents a repulsive force of the obstacle i, satisfying the following formula:

wherein,

expression η _i A gradient;

step 3-2) establishing a gravitational field by combining the conditions of the front and rear vehicles of the flying vehicle to be planned by decision;

gravitational field U acting on aerocar _a (X) and gravitational force F _a (X) satisfies the following formula:

where ρ represents a gravitational direct proportionality coefficient, X _g Representing the target position, η (X, X) _g ) Indicating the distance between the flying car and the target location,

representing the η gradient.

5. The flying automobile takeoff and landing decision planning method based on three-dimensional target detection as claimed in claim 1, wherein the step 4) is to determine whether the distance between the flying automobile and the target position meets a safe distance; the method comprises the following steps:

and when the distance between the front vehicle and the rear vehicle is greater than or equal to the safety distance, the safety distance is met, otherwise, the safety distance is not met.

6. The flying vehicle takeoff and landing decision-making planning method based on three-dimensional target detection as claimed in claim 1, wherein the step 5) comprises:

step 5-1) extracting the data of the automobile running on a one-way road and the flight data of the aerocar to be planned by decision, and establishing a repulsive field after analysis and processing;

repulsive force field U acting on aerocar _ri (X) and repulsive force F _r (X) satisfies the following formula:

wherein k is _r Expressing a direct proportionality coefficient, X expressing the position of the hovercar to be planned by decision, X _i Indicates the position, η, of the obstacle i _i (X,X _i ) Denotes X and X _i A distance therebetween, η _i (X,X _i ) Denotes X and X _i A distance therebetween, η _0，i Represents the maximum distance over which the ith obstacle repulsive field acts;

step 5-2) selecting an air safety position to establish a gravitational field;

gravitational field U acting on aerocar _a (X) and gravitational force F _a (X) satisfies the following formula:

where ρ represents a gravitational direct proportionality coefficient, X _g Representing the target position, η (X, X) _g ) Indicating the distance between the flying car and the target location,

representing the η gradient.

7. A flying car take-off and landing decision-making planning system based on three-dimensional target detection is realized based on a camera and a laser radar which are deployed on a flying car, and is characterized by comprising the following components: the system comprises a dynamic three-dimensional map building module, a state judging module, a flight repulsive gravitational field building module, a landing control module, a driving repulsive gravitational field building module and a take-off control module;

the dynamic three-dimensional map building module is used for sensing the environment of the specified road according to the image and the three-dimensional point cloud data collected in real time and building a dynamic three-dimensional map;

the state judgment module is used for switching to a flight repulsive gravitational field establishment module when the flying automobile to be planned is in a flight state; when the flying automobile to be planned is in a driving state, the flying automobile is switched to a driving repulsive gravitational field establishing module;

the flight repulsive gravitational field establishing module is used for establishing a repulsive field according to the dynamic three-dimensional map by combining the data of the flying automobile and the condition of the automobile running on a one-way road, and selecting a certain point behind the running automobile or a certain point on the ground to establish a gravitational field;

the landing control module is used for forming a real-time changing resultant force field by the repulsive force field and the gravitational field, driving the flying automobile to dynamically land, judging whether the distance between the flying automobile and the target position meets the safety distance, if so, controlling the flying automobile to land, and finishing decision planning; if not, controlling the flying automobile to ascend, and transferring to a flying repulsive gravitational field establishing module to reestablish a gravitational field;

the driving repulsive gravitational field establishing module is used for establishing a repulsive gravitational field according to the dynamic three-dimensional map by combining the flying automobile data and the automobile condition driven on a one-way road, and selecting an air safety position to establish a gravitational field;

and the takeoff control module is used for forming a real-time changing resultant force field by the repulsion field and the gravitational field, driving the aerocar to take off and finishing decision planning.

Images