CN-121998201-A - Agricultural machinery operation obstacle early warning method based on real-time environment information

CN121998201ACN 121998201 ACN121998201 ACN 121998201ACN-121998201-A

Abstract

An agricultural machinery operation obstacle early warning method based on real-time environment information relates to the technical field of intelligent safety control of agricultural machinery. The existing agricultural machinery obstacle early warning technology has inaccurate early warning. According to the position of the agricultural machine acquired in real time and n obstacle positions in a preset range, the Euclidean distance between the agricultural machine and each obstacle and the area where each obstacle is located relative to the heading of the agricultural machine are calculated in real time, a preset dynamic early warning distance threshold corresponding to each obstacle is obtained according to the distance, the current operation mode of the agricultural machine and each obstacle category, the normalized distance ratio of each obstacle is calculated according to the Euclidean distance between the agricultural machine and each obstacle and the corresponding preset dynamic early warning distance threshold, so that corresponding risk values are obtained, corresponding risk grades are obtained according to each risk value, no alarm is carried out when the risk grades are all 0, and when the risk grades are not all 0, the alarm is carried out. The invention is used for barrier early warning.

Inventors

WANG YIJIA
Yan Jiwenqing
GAO FENG
WANG NAIHUI

Assignees

东北农业大学

Dates

Publication Date: 20260508
Application Date: 20260211

Claims (9)

1. An agricultural machinery operation obstacle early warning method based on real-time environment information is characterized by comprising the following steps: step1, calculating the Euclidean distance between the agricultural machinery and each obstacle and the area where each obstacle is located relative to the agricultural machinery heading in real time according to the agricultural machinery position acquired in real time and n obstacle positions in a preset range; Step 2, obtaining a preset dynamic early warning distance threshold value corresponding to each obstacle according to the Euclidean distance between the agricultural machinery and each obstacle, the current agricultural machinery operation mode and each obstacle category; And 3, calculating the normalized distance ratio of each obstacle according to the Euclidean distance between the agricultural machine and each obstacle and the corresponding preset dynamic early warning distance threshold value, so as to obtain corresponding risk values, obtaining corresponding risk levels according to each risk value, not alarming when the risk levels are all 0, alarming when the risk levels are not all 0, selecting obstacle information corresponding to the maximum value of the risk levels from the obtained risk levels for outputting, wherein the obstacle information comprises the area where the obstacle is relative to the agricultural machine course, the type of the obstacle and the Euclidean distance between the agricultural machine and each obstacle, and selecting the obstacle information corresponding to the minimum value of the Euclidean distance between the agricultural machine and each obstacle for outputting when a plurality of the same maximum risk levels exist.
2. The method for early warning of an obstacle in operation of an agricultural machine based on real-time environmental information according to claim 1, wherein the euclidean distance between the agricultural machine and each obstacle is: , in the formula, Is the agricultural machinery and the first The euclidean distance between the individual obstacles, Is the position and the first of the agricultural machinery The relative position vector of the individual obstacles, , Is the first The position of the individual obstacle(s), Is used as a position of the agricultural machinery, , , Is the abscissa of the agricultural machinery and is the coordinate, Is the ordinate of the agricultural machinery, Is the first The abscissa of the individual obstacles is taken to mean, Is the first The abscissa of the individual obstacles.
3. The method for early warning of an obstacle in agricultural machinery operation based on real-time environmental information according to claim 1 or 2, wherein the process of calculating the area where each obstacle is located relative to the heading of the agricultural machinery is as follows: the angle of each obstacle relative to the agricultural machine heading is: , in the formula, Is the first The angle of the individual obstacles relative to the heading of the agricultural machine, For obstacles relative to the global coordinate system The absolute azimuth angle of the shaft, For the agricultural machinery to the global coordinate system The heading angle of the axis, ; Preset direction boundary The method comprises the following steps of: , in the formula, Is the first A number of pre-set directional boundaries are provided, Indicating that the obstacle is relative to the area directly in front of the agricultural machine heading, Indicating that the obstacle is navigating to the left front area relative to the agricultural machine, Representing the area to the right and left of the obstacle relative to the agricultural machine heading, Indicating that the obstacle is navigating to the rear left relative to the agricultural machine, Indicating that the obstacle is relative to the area directly behind the heading of the agricultural machine, Indicating that the obstacle is navigating to the rear right area relative to the agricultural machine, Representing the right square area of the obstacle relative to the heading of the agricultural machine, Representing the right front region of the obstacle relative to the heading side of the agricultural machine.
4. The method for early warning of an obstacle in operation of an agricultural machine based on real-time environmental information as claimed in claim 3, wherein, according to the Euclidean distance between the agricultural machinery and each obstacle, the current agricultural machinery operation mode and each obstacle category, a preset dynamic early warning distance threshold value corresponding to each obstacle is obtained, and the specific process is as follows: Step 21, presetting early warning parameter values corresponding to different agricultural machinery operation modes, obstacle categories and combination modes of areas where obstacles are located relative to agricultural machinery heading, wherein the early warning parameter values comprise basic early warning distance, speed-related correction coefficients and braking safety margin parameters; Step 22, selecting basic early warning distance, speed related correction coefficient and braking safety margin parameters from preset early warning parameter values according to the current agricultural machinery operation mode, obstacle category and area where the obstacle is located relative to the agricultural machinery heading; Step 23, acquiring the linear speed of the agricultural machinery in real time, presetting the response time of a driver and presetting the maximum braking deceleration according to the model of the agricultural machinery; And step 24, obtaining a dynamic early warning distance threshold of each obstacle according to the selected basic early warning distance, speed related correction coefficient, agricultural machinery linear speed, preset driver response time, preset maximum braking deceleration, basic early warning distance, speed related correction coefficient and braking safety margin parameters.
5. The method for early warning an obstacle in agricultural machinery operation based on real-time environmental information according to claim 4, wherein the dynamic early warning distance threshold function of the obstacle is: , in the formula, Is the first The individual obstacle is at the moment Is used for the dynamic early warning distance threshold value, Is the linear speed of the agricultural machinery, In order to be a physical safety distance, , In order to achieve the reaction distance, the reaction time, In order to be able to brake the distance, In order to preset the response time of the driver, In order to preset the maximum braking deceleration rate, , , Is the first The velocity-dependent correction coefficients of the individual obstacles, Is the first The basic pre-warning distance of the individual obstacles, Is the first And a brake safety margin parameter for each obstacle.
6. The method for early warning of an obstacle in agricultural machinery operation based on real-time environmental information according to claim 5, wherein the normalized distance ratio of the obstacle is The method comprises the following steps: , Risk value The method comprises the following steps: 。
7. the method for early warning the obstacle in the operation of the agricultural machinery based on the real-time environment information according to claim 6, wherein the risk level is obtained according to the risk value correspondence, specifically: Step 31, presetting a risk level dividing constraint condition; Step 32, selecting risk level classification parameters from preset risk level classification constraint conditions according to the current agricultural machine operation mode, the obstacle category and the area where the obstacle is located relative to the agricultural machine course; and step 33, establishing a risk level according to the risk value and the risk level dividing parameter.
8. The method for early warning of an obstacle in operation of an agricultural machine based on real-time environmental information as claimed in claim 7, wherein the risk classification constraint condition is: , in the formula, And Respectively are in operation modes Class of obstacle And the area where the obstacle is located relative to the heading of the agricultural machine A preset low risk and medium risk demarcation threshold in combination and a medium risk and high risk demarcation threshold.
9. The method for early warning of an obstacle in operation of an agricultural machine based on real-time environmental information as claimed in claim 8, wherein the risk level is The method comprises the following steps: , in the formula, Representing no early warning; representing low-level early warning; Representing intermediate early warning; representing advanced warning.

Description

Agricultural machinery operation obstacle early warning method based on real-time environment information Technical Field The invention relates to the technical field of intelligent safety control of agricultural machinery. Background With the improvement of the power and automation level of agricultural machinery, agricultural machinery such as tractors, combine harvesters and the like continuously work for a long time in a field environment, and various obstacles such as pedestrians, livestock, other agricultural machinery, ditches, tree piles and the like exist around the agricultural machinery. Once the early warning is not timely or the false warning is frequent, the safety accident or the operation efficiency is possibly reduced. The existing agricultural machinery obstacle early warning technology generally adopts a mode of single fixed threshold value or linear adjustment along with the speed of a vehicle to alarm the detected obstacle. Such methods suffer from the following disadvantages: the difference of operation modes, namely, the dangerous degree and the allowable approaching distance of obstacles around the agricultural machinery are different in different operation modes such as transition high-speed running, low-speed sowing, pesticide spraying, harvesting and the like, is not fully considered, and the early warning strategy is not dynamically changed when the mode is switched in the existing method. The system is not supported by a mathematical model, and a plurality of systems only set the alarm distance and the alarm level through experience, and lack of a unified mathematical modeling frame, so that the subsequent parameter calibration and system expansion are not facilitated. The early warning level is not fine enough, and the grading early warning information cannot be given according to the operation scene and the obstacle type by adopting binary judgment of alarm/not alarm. Therefore, the existing agricultural machinery obstacle early warning technology has the problem of inaccurate early warning. Disclosure of Invention The invention aims to solve the problem of inaccurate early warning in the existing agricultural machinery obstacle early warning technology, and provides an agricultural machinery operation obstacle early warning method based on real-time environment information. An agricultural machinery operation obstacle early warning method based on real-time environment information, the method comprises the following steps: step1, calculating the Euclidean distance between the agricultural machinery and each obstacle and the area where each obstacle is located relative to the agricultural machinery heading in real time according to the agricultural machinery position acquired in real time and n obstacle positions in a preset range; Step 2, obtaining a preset dynamic early warning distance threshold value corresponding to each obstacle according to the Euclidean distance between the agricultural machinery and each obstacle, the current agricultural machinery operation mode and each obstacle category; And 3, calculating the normalized distance ratio of each obstacle according to the Euclidean distance between the agricultural machine and each obstacle and the corresponding preset dynamic early warning distance threshold value, so as to obtain corresponding risk values, obtaining corresponding risk levels according to each risk value, not alarming when the risk levels are all 0, alarming when the risk levels are not all 0, selecting obstacle information corresponding to the maximum value of the risk levels from the obtained risk levels for outputting, wherein the obstacle information comprises the area where the obstacle is relative to the agricultural machine course, the type of the obstacle and the Euclidean distance between the agricultural machine and each obstacle, and selecting the obstacle information corresponding to the minimum value of the Euclidean distance between the agricultural machine and each obstacle for outputting when a plurality of the same maximum risk levels exist. Preferably, the euclidean distance between the agricultural machine and each obstacle is: , in the formula, Is the agricultural machinery and the firstThe euclidean distance between the individual obstacles,Is the position and the first of the agricultural machineryThe relative position vector of the individual obstacles,,Is the firstThe position of the individual obstacle(s),Is used as a position of the agricultural machinery,,,Is the abscissa of the agricultural machinery and is the coordinate,Is the ordinate of the agricultural machinery,Is the firstThe abscissa of the individual obstacles is taken to mean,Is the firstThe abscissa of the individual obstacles. Preferably, the process of calculating the area where each obstacle is located relative to the heading of the agricultural machine is as follows: the angle of each obstacle relative to the agricultural machine heading is: , in the formula