Traveling obstacle avoidance algorithm based on four-dimensional theory

Abstract

The invention relates to the technical field of data processing, in particular to a four-dimensional theory-based advancing obstacle avoidance algorithm, which comprises the following steps of: sensing obstacle target contour information in an advancing direction based on environment sensing equipment, and constructing an obstacle virtual four-dimensional model based on the sensed obstacle target contour information; setting a cloud database, uploading the contrast model to the cloud database for storage, obtaining the constructed obstacle virtual four-dimensional model, and searching for similar contrast models in the cloud database; the obstacle virtual four-dimensional model avoidance emergency degree is analyzed, and an obstacle avoidance queue is further generated based on the obstacle virtual four-dimensional model avoidance emergency degree; according to the invention, the algorithm greatly improves the efficiency of data processing, can obtain an accurate result in a shorter time regardless of a large-scale data set or a complex calculation task, and saves a large amount of time cost for enterprises and scientific research institutions.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a traveling obstacle avoidance algorithm based on four-dimensional theory. Background Art

[0002] Obstacle avoidance refers to automatically detecting and avoiding obstacles during movement to ensure safe and efficient arrival at the target location. Sensors such as lidar, cameras, and ultrasound are usually used to sense the surrounding environment. When an obstacle is detected, the algorithm analyzes its position, size, motion state, and other information to plan the best path. This is crucial in the fields of robot navigation and autonomous driving. It enables equipment to flexibly shuttle in complex environments, improves work efficiency and safety, and brings great convenience to people's production and life. It is one of the important directions of modern scientific and technological development.

[0003] The invention patent with application number 202210398942.4 discloses an automatic obstacle avoidance method based on computer vision, which is characterized by comprising the following steps: step S100: setting the shooting angle of the camera (1) and the position and detection angle of the laser radar (2) on the moving device; step S200: initializing the camera (1) and the laser radar (2); step S300: jointly calibrating the coordinate system of the camera (1) and the laser radar (2); step S400: calculating the current moving speed and position information of the moving device according to the preset distance measurement rules; step S500: calculating the current moving speed and position information of the moving device according to the preset distance measurement rules; step S600: calculating the current moving speed and position information of the moving device according to the preset distance measurement rules; step S700: calculating the current moving speed and position information of the moving device according to the preset distance measurement rules; step S800: calculating the current moving speed and position information of the moving device according to the preset distance measurement rules; step S900: calculating the current moving speed and position information of the moving device according to the preset distance measurement rules; step S1 ... 00: Determine whether there is an obstacle in the direction of travel: If yes, go to step S600: If not, go to step S700; Step S600: Change the planned travel path according to the preset obstacle avoidance rules: Step S700: Control the travel direction and travel speed according to the planned travel path: The step S400 includes: Step S401: Obtain the position of the sampling point and the ranging point closest to the traveling device within the travel direction range: Step S402: Calculate the travel speed and the position information based on the relative position change between the ranging point and the sampling point and the traveling device.

[0004] This application aims to solve the problem that "agricultural machinery needs to avoid obstacles when operating automatically in the fields. The traditional obstacle avoidance method uses the shortest tangent method. This method generates a fast, simple and short trajectory, but the curvature of the shortest path curve is discontinuous and there is a bending point at the connection between the straight line and the arc. This will cause the front wheel angle to change suddenly at the beginning of the turn and when passing an obstacle and returning to a straight line, resulting in unstable mechanical operation. Moreover, the cameras and ranging sensors used in obstacle avoidance are easily installed in different positions, which may cause deviations in obstacle avoidance. In addition, the existing obstacle avoidance methods will avoid obstacles indiscriminately, regardless of the obstacle encountered. Therefore, it greatly affects the obstacle avoidance driving accuracy of agricultural machinery and reduces the success rate of the obstacle avoidance process."

[0005] However, obstacle avoidance technology is also used in the field of intelligent assisted driving of automobiles, but currently there is no related technology for designing obstacle avoidance priority based on the urgency of obstacles in intelligent assisted driving of automobiles.

[0006] To this end, we propose a moving obstacle avoidance algorithm based on four-dimensional theory. Summary of the invention

[0007] In view of the above-mentioned shortcomings of the prior art, the present invention provides a moving obstacle avoidance algorithm based on four-dimensional theory, which solves the technical problems raised in the above-mentioned background technology.

[0008] To achieve the above objectives, the present invention is implemented through the following technical solutions: A moving obstacle avoidance algorithm based on four-dimensional theory, including: Based on the obstacle target contour information perceived by the environmental perception device in the direction of travel, a virtual four-dimensional model of the obstacle is constructed based on the perceived obstacle target contour information; a cloud database is set, a reference model is uploaded to the cloud database for storage, the constructed virtual four-dimensional model of the obstacle is obtained, and a similar reference model is searched in the cloud database; the urgency of avoiding the virtual four-dimensional model of the obstacle is analyzed, and an obstacle avoidance queue is generated based on the urgency of avoiding the virtual four-dimensional model of the obstacle; obstacles are avoided in sequence according to the obstacle avoidance queue, obstacle avoidance images are collected in real time, and corrections are set according to the obstacle avoidance images; and the configuration weight of the virtual four-dimensional model of the obstacle is corrected according to the set corrections.

[0009] Furthermore, the environmental perception device includes an ultrasonic sensor, a ranging sensor and a lidar. The obstacle target contour information perceived by the environmental perception device includes: the distance, shape and distribution space area from the obstacle target. The perception distance and perception angle of the environmental perception device are customized by the user.

[0010] Furthermore, after the virtual four-dimensional model of the obstacle is constructed, it is placed in the same space according to the relative posture of each model based on the obstacle target contour information used for model construction. After the placement is completed, the space where each group of virtual four-dimensional models of obstacles are placed is recorded as an obstacle virtual scene.

[0011] Furthermore, the reference models stored in the cloud database are manually uploaded by the user end, and the reference models include: buildings, animals, roadblocks, vehicles, and plants. When searching for similar reference models in the cloud database, analysis is performed based on any algorithm of geometric features, topological results, feature descriptors, deep learning, or space occupancy.

[0012] Furthermore, when the similarity between the reference model and the virtual four-dimensional model of the obstacle is greater than 95%, the reference model similar to the virtual four-dimensional model of the obstacle is determined to be a similar reference model found in the cloud database.

[0013] Furthermore, when the virtual four-dimensional model of the obstacle does not find a similar target in the cloud database, the virtual four-dimensional model of the obstacle is extracted from the virtual scene of the obstacle, and synchronously sent to the cloud database and stored in the cloud database.

[0014] Furthermore, the virtual four-dimensional models of obstacles in the virtual obstacle scene are all set and configured by the user end to distinguish weights, and the urgency of the virtual four-dimensional models of obstacles is analyzed based on the distinguishing weights of the virtual four-dimensional models of obstacles; The emergency analysis logic of the virtual four-dimensional model of obstacles is expressed as follows: ; Where: The urgency of creating a virtual 4D model of the obstacle; is the configuration weight of the virtual four-dimensional model of the obstacle; is the distance from the obstacle to the sensing device.

[0015] Furthermore, the configuration weight of the virtual four-dimensional model of the obstacle The value of is subject to: The obstacle virtual four-dimensional model corresponds to the obstacle type of buildings, animals, roadblocks, vehicles, plants, recorded as 1. 2. 3. 4. 5. 1> 4> 5> 2> 3; The virtual four-dimensional model of the obstacle in the obstacle avoidance queue is based on the urgency of the virtual four-dimensional model of the obstacle. Sort in descending order.

[0016] Furthermore, the obstacle corresponding correction is in the value range of 0 to 1, and the correction operation on the virtual four-dimensional model of the obstacle is the product of the configuration weight of the virtual four-dimensional model of the obstacle and the correction; Among them, the obstacle corresponding correction value follows: the greater the braking distance with the obstacle before the avoidance operation in the obstacle avoidance image, the smaller the obstacle corresponding correction value, and vice versa.

[0017] Compared with the known prior art, the technical solution provided by the present invention has the following beneficial effects: The present invention provides a traveling obstacle avoidance algorithm based on four-dimensional theory. During the execution of the algorithm, the efficiency of data processing is greatly improved. Whether it is a large-scale data set or a complex computing task, accurate results can be obtained in a shorter time, saving a lot of time costs for enterprises and scientific research institutions. Secondly, the algorithm has high accuracy and reliability. Through precise calculation and optimization, the probability of error is reduced, providing solid data support for decision-making. Furthermore, it has good adaptability and scalability, and can easily cope with problems in different fields and scales, and can be easily upgraded and improved as data continues to increase and problems change. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0019] Figure 1 The figure is a flow chart of a moving obstacle avoidance algorithm based on four-dimensional theory. DETAILED DESCRIPTION

[0020] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0021] The present invention will be further described below in conjunction with the embodiments.

[0022] Embodiment 1: A traveling obstacle avoidance algorithm based on four-dimensional theory in this embodiment, such as Figure 1 As shown, the following steps are included: Step 1: Based on the environmental perception device, the obstacle target contour information in the direction of travel is perceived, and a virtual four-dimensional model of the obstacle is constructed based on the perceived obstacle target contour information; Step 2: Set up a cloud database, upload the reference model to the cloud database for storage, obtain the constructed virtual 4D model of the obstacle, and search for similar reference models in the cloud database; Step 3: Analyze the urgency of avoiding the virtual four-dimensional model of obstacles, and further generate an obstacle avoidance queue based on the urgency of avoiding the virtual four-dimensional model of obstacles; Step 4: Avoid obstacles in order according to the obstacle avoidance queue, collect obstacle avoidance images in real time, and set corrections based on the obstacle avoidance images; Step 5: Correct the configuration weight of the virtual four-dimensional model of the obstacle according to the set correction.

[0023] like Figure 1 As shown, the environmental perception device includes an ultrasonic sensor, a ranging sensor and a lidar. The obstacle target contour information perceived by the environmental perception device includes: the distance to the obstacle target, the shape and the distribution space area. The perception distance and perception angle of the environmental perception device are customized by the user.

[0024] In this embodiment, by executing the steps in the above embodiments, the efficiency of data processing is greatly improved, providing a safer and more reliable obstacle avoidance service for intelligent assisted driving of automobiles.

[0025] Embodiment 2: In terms of specific implementation, based on Example 1, this example refers to Figure 1 A traveling obstacle avoidance algorithm based on four-dimensional theory in Example 1 is further specifically described as follows: After the obstacle virtual four-dimensional model is built, it is placed in the same space according to the relative posture of each model based on the obstacle target contour information used for model construction. After the placement is completed, the space where each group of obstacle virtual four-dimensional models is placed is recorded as an obstacle virtual scene.

[0026] like Figure 1 As shown, the reference models stored in the cloud database are manually uploaded by the user. The reference models include: buildings, animals, roadblocks, vehicles, and plants. When searching for similar reference models in the cloud database, analysis is performed based on any algorithm of geometric features, topological results, feature descriptors, deep learning, or space occupancy.

[0027] like Figure 1 As shown, when the virtual four-dimensional model of the obstacle does not find a similar target in the cloud database, the virtual four-dimensional model of the obstacle is extracted from the virtual scene of the obstacle, and synchronously sent to the cloud database and stored in the cloud database.

[0028] In this embodiment, through the above settings, further step execution data and logic support are provided for the step execution of the algorithm in Example 1, ensuring stable, reliable and effective execution of the algorithm in the embodiment.

[0029] Embodiment 3: In terms of specific implementation, based on Embodiment 1, this embodiment refers to Figure 1 A traveling obstacle avoidance algorithm based on four-dimensional theory in Example 1 is further specifically described as follows: The virtual four-dimensional models of obstacles in the virtual scene of obstacles are all set and configured by the user end to distinguish weights, and the urgency of the virtual four-dimensional models of obstacles is analyzed based on the distinguishing weights of the virtual four-dimensional models of obstacles; The logic of the urgency analysis of the virtual four-dimensional obstacle model is expressed as follows: ; Where: The urgency of creating a virtual 4D model of the obstacle; is the configuration weight of the virtual four-dimensional model of the obstacle; is the distance from the obstacle to the sensing device.

[0030] like Figure 1 As shown in the figure, the configuration weight of the virtual four-dimensional model of the obstacle The value of is subject to: The obstacle virtual four-dimensional model corresponds to the obstacle type of buildings, animals, roadblocks, vehicles, plants, recorded as 1. 2. 3. 4. 5. 1> 4> 5> 2> 3; Among them, the virtual four-dimensional model of obstacles in the obstacle avoidance queue is based on the urgency of the virtual four-dimensional model of obstacles. Sort in descending order.

[0031] like Figure 1 As shown, the obstacle corresponding correction is in the value range of 0~1, and the correction operation of the virtual four-dimensional model of the obstacle is the product of the configuration weight of the virtual four-dimensional model of the obstacle and the correction; Among them, the obstacle corresponding correction value follows: the greater the braking distance with the obstacle before the avoidance operation in the obstacle avoidance image, the smaller the obstacle corresponding correction value, and vice versa.

[0032] In this embodiment, through the above settings, further step execution data and logic support are provided for the step execution of the algorithm in Example 1, and the urgency analysis logic based on the virtual four-dimensional model of the obstacle is further limited.

[0033] In summary, during the execution of the algorithm in the above embodiment, the efficiency of data processing is greatly improved. Whether it is a large-scale data set or a complex computing task, accurate results can be obtained in a shorter time, saving a lot of time costs for enterprises and scientific research institutions; secondly, the algorithm has a high degree of accuracy and reliability. Through precise calculation and optimization, the probability of errors is reduced, providing solid data support for decision-making; thirdly, it has good adaptability and scalability, and can easily cope with problems in different fields and scales, and can be easily upgraded and improved as data continues to increase and problems change.

[0034] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A moving obstacle avoidance algorithm based on four-dimensional theory, characterized in that: The following steps are involved: Step 1: Based on the environmental perception device, the obstacle target contour information in the direction of travel is perceived, and a virtual four-dimensional model of the obstacle is constructed based on the perceived obstacle target contour information; Step 2: Set up a cloud database, upload the reference model to the cloud database for storage, obtain the constructed virtual 4D model of the obstacle, and search for similar reference models in the cloud database; Step 3: Analyze the urgency of avoiding the virtual four-dimensional model of obstacles, and further generate an obstacle avoidance queue based on the urgency of avoiding the virtual four-dimensional model of obstacles; Step 4: Avoid obstacles in order according to the obstacle avoidance queue, collect obstacle avoidance images in real time, and set corrections based on the obstacle avoidance images; Step 5: Correct the configuration weight of the virtual four-dimensional model of the obstacle according to the set correction.

2. The obstacle avoidance algorithm based on four-dimensional theory according to claim 1, characterized in that: The environmental perception device includes an ultrasonic sensor, a ranging sensor and a laser radar. The obstacle target contour information perceived by the environmental perception device includes: the distance, shape and distribution space area to the obstacle target. The perception distance and perception angle of the environmental perception device are customized by the user.

3. The obstacle avoidance algorithm based on four-dimensional theory according to claim 1, characterized in that: After the obstacle virtual four-dimensional model is constructed, the obstacle target contour information used for model construction is synchronously placed in the same space according to the relative posture of each model. After the placement is completed, the space where each group of obstacle virtual four-dimensional models is placed is recorded as an obstacle virtual scene.

4. The obstacle avoidance algorithm based on four-dimensional theory according to claim 1, characterized in that: The reference models stored in the cloud database are manually uploaded by the user end, and the reference models include: buildings, animals, roadblocks, vehicles, and plants. When searching for similar reference models in the cloud database, analysis is performed based on any algorithm including geometric features, topological results, feature descriptors, deep learning, or space occupancy.

5. The obstacle avoidance algorithm based on four-dimensional theory according to claim 4, characterized in that: When the similarity between the reference model and the virtual four-dimensional model of the obstacle is greater than 95%, the reference model similar to the virtual four-dimensional model of the obstacle is determined to be a similar reference model found in the cloud database.

6. The obstacle avoidance algorithm based on four-dimensional theory according to claim 4, characterized in that: When the virtual four-dimensional model of the obstacle does not find a similar target in the cloud database, the virtual four-dimensional model of the obstacle is extracted from the virtual scene of the obstacle, and synchronously sent to the cloud database and stored in the cloud database.

7. The obstacle avoidance algorithm based on four-dimensional theory according to claim 1, characterized in that: The virtual four-dimensional models of obstacles in the virtual obstacle scene are all set and configured by the user end to distinguish weights, and the urgency of the virtual four-dimensional models of obstacles is analyzed based on the distinguishing weights of the virtual four-dimensional models of obstacles; The emergency analysis logic of the virtual four-dimensional model of obstacles is expressed as follows: ； Where: The urgency of creating a virtual 4D model of the obstacle; Configuration weight of virtual 4D model of obstacle; is the distance from the obstacle to the sensing device.

8. The obstacle avoidance algorithm based on four-dimensional theory according to claim 7, characterized in that: Configuration weight of the virtual four-dimensional model of the obstacle The value of is subject to: The obstacle virtual four-dimensional model corresponds to the obstacle type of buildings, animals, roadblocks, vehicles, plants, recorded as 1.

2.

3.

4.

5. 1> 4> 5> 2> 3; The virtual four-dimensional model of the obstacle in the obstacle avoidance queue is based on the urgency of the virtual four-dimensional model of the obstacle. Sort in descending order.

9. The obstacle avoidance algorithm based on four-dimensional theory according to claim 1, characterized in that: The obstacle corresponding correction is in the value range of 0 to 1, and the correction operation on the virtual four-dimensional model of the obstacle is the product of the configuration weight of the virtual four-dimensional model of the obstacle and the correction; Among them, the obstacle corresponding correction value follows: the greater the braking distance with the obstacle before the avoidance operation in the obstacle avoidance image, the smaller the obstacle corresponding correction value, and vice versa.

Images