CN117629185B - A photovoltaic cleaning robot operation path planning method under complex working conditions - Google Patents

Abstract

The invention discloses a photovoltaic cleaning robot running path planning method under a complex working condition, and belongs to the technical field of path planning. According to the invention, the boundary searching strategy and the bottom searching strategy are introduced, so that the searching range of the photovoltaic cleaning robot in operation is enlarged, and the bottom searching strategy ensures that the photovoltaic cleaning robot can effectively return to the previous area after entering a new area and completing the cleaning task of the whole area, so that the unfinished part is continuously cleaned, the efficiency and the adaptability of the photovoltaic cleaning robot are improved, and the photovoltaic cleaning robot can better cope with the challenges of complex environments.

Description

A photovoltaic cleaning robot operation path planning method under complex working conditions

Technical Field

The present invention relates to the technical field of path planning, and in particular to a method for planning the operation path of a photovoltaic cleaning robot under complex working conditions.

Background technique

The operation logic of photovoltaic cleaning robots is currently restricted by many factors, mainly including the use of ordinary line wrapping, ATK positioning technology and the SLAM method of traditional cleaning robots. These methods have defects in different aspects, especially in specific background environments, their operation may be limited.

Ordinary line wrapping: Ordinary line wrapping methods are usually the initial choice for photovoltaic cleaning robots because they are relatively simple and low-cost. However, the applicability of this method is greatly limited in complex outdoor environments. Photovoltaic panels are often distributed in large outdoor areas with various terrains, obstacles and uneven surfaces. Ordinary line wrapping cannot cope with these complex terrains and environments, resulting in the cleaning robot being unable to complete the task effectively and easily getting stuck in obstacles or getting into trouble.

ATK positioning technology: Although ATK (Absolute Track and Know) positioning technology has improved the positioning accuracy of photovoltaic cleaning robots to a certain extent, it still has some limitations. First, ATK relies on specific types of signs or markers for positioning, which means that these markers need to be arranged on solar panels, which may not be practical in practice. In addition, obstacle occlusion and positioning accuracy issues still exist, because the ATK system may not be able to effectively identify or track markers, especially in a constantly changing outdoor environment.

SLAM method of traditional cleaning robots: Traditional cleaning robots use SLAM (Simultaneous Localization and Mapping) technology to capture the surrounding environment through cameras to locate themselves and plan cleaning paths. However, this method has major flaws in the field of outdoor photovoltaic cleaning. The lighting and weather conditions in the outdoor environment may have a negative impact on the performance of the camera, thereby affecting the accuracy of positioning. In addition, the instability and diversity in the outdoor environment make it difficult for the SLAM method to cope with complex photovoltaic panel layouts.

In the current context, in order to overcome the above problems, it is necessary to find new operation logic and technical solutions to improve the efficiency and adaptability of photovoltaic cleaning robots. To this end, a photovoltaic cleaning robot operation path planning method under complex working conditions is proposed.

Summary of the invention

The technical problem to be solved by the present invention is: how to solve the path planning problem after connecting multiple randomly distributed areas, and provides a photovoltaic cleaning robot operation path planning method under complex working conditions. By more reasonably planning the operation path of the photovoltaic cleaning robot, the efficiency and adaptability of the photovoltaic cleaning robot are improved, and it is more suitable for photovoltaic panel cleaning work under complex working conditions.

The present invention solves the above technical problems through the following technical solutions, and the present invention comprises the following steps:

S1: At any boundary position of the cleaning task execution area as the initial position, the photovoltaic cleaning robot determines the boundary of the initial position as the top and sets the initial forward direction as the right;

S2: After starting to clean, the photovoltaic cleaning robot executes the line-changing strategy and moves downward line by line. When the photovoltaic cleaning robot is running, the photovoltaic cleaning robot continuously monitors the feedback of the left radar and obtains the detection distance value of the left radar in real time. When the detection distance value of the left radar decreases from exceeding the set distance value to within the set distance value, it indicates that a new area appears on the left. The boundary-finding strategy is executed, and the robot turns left and enters the new area.

S3: The photovoltaic cleaning robot turns left and enters a new area and continues to run. During the running process, it continues to execute the line-breaking strategy and the boundary-finding strategy, and follows the bottom-finding strategy until the photovoltaic cleaning robot returns to the initial position and ends the cleaning task.

Furthermore, in the steps S1 to S3, when the photovoltaic cleaning robot moves along the boundary, the left radar position of the photovoltaic cleaning robot is outside the boundary, and the detection distance value exceeds the set value.

Furthermore, the left radar, front radar and right radar are all installed at the lower end of the photovoltaic cleaning robot, and all detect in the direction perpendicular to the plane where the photovoltaic panels are located, wherein the front radar is located in front of the left radar and the right radar, and the left radar and the right radar are located on the same straight line perpendicular to the axis of the photovoltaic cleaning robot, and the straight line is parallel to the plane where the photovoltaic cleaning robot is located.

Furthermore, in steps S2 and S3, during the operation of the photovoltaic cleaning robot, it is determined whether the current position is a left corner or a right corner based on the running direction and the relationship between the detection distance values of the front radar, the left radar, and the right radar and the set distance values.

Furthermore, in step S2, the line-changing strategy is as follows: when the photovoltaic cleaning robot reaches the left and right boundaries of the photovoltaic panel, the robot performs a downward line-changing action and then runs in the opposite direction to continue the cleaning task.

Furthermore, in the step S3, when executing the boundary-seeking strategy, it is necessary to follow the bottom-probing strategy, which is as follows: when the photovoltaic cleaning robot is cleaning the current area, if it has not reached the right corner or the left corner and has not obtained the opportunity to explore the bottom, and is currently driving in a straight line to the left, when the detection distance value of the left radar changes from exceeding the set distance value to being within the set distance value, it indicates that a new area has appeared below the front. Since the photovoltaic cleaning robot has not reached the left and right corners at the bottom of the current area, it cannot enter the new area below the front in advance, and the photovoltaic cleaning robot performs a U-turn and returns.

Furthermore, the triggering method of the bottom exploration opportunity is as follows: when the photovoltaic cleaning robot reaches the left or right corner during operation, it will get a bottom exploration opportunity. During the photovoltaic cleaning robot's operation to the left, if the detection distance value of the left radar changes from within the set distance value to beyond the set distance value, the photovoltaic cleaning robot will continue to move forward; during the operation to the left, if the detection distance value of the left radar changes from exceeding the set distance value to within the set distance value, it indicates that there is a lower area ahead that needs to be explored downward; at this time, the photovoltaic cleaning robot with the bottom exploration opportunity will turn left, enter the lower area, and execute the line break strategy, advancing downward line by line.

Furthermore, the left corner is the junction of the lower boundary and the left boundary of the area above the lower boundary, and the right corner is the junction of the lower boundary and the right boundary of the area above the lower boundary; wherein, taking the initial position as a reference, when the bottom of any photovoltaic panel is not connected to other photovoltaic panels, the bottom of the photovoltaic panel is the lower boundary or part of the lower boundary.

Furthermore, when the photovoltaic cleaning robot completes a downward movement, the bottom exploration opportunity will be cancelled. The downward movement includes executing the downward movement in the line change strategy, but does not include a U-turn.

Furthermore, in step S3, the initial position area is set to be all white or all black. When the photovoltaic cleaning robot reaches the initial position area, the characteristic information is recognized by multiple diffuse reflection sensors set at its lower end, and the current position can be confirmed as the initial position, thereby terminating the operation of the photovoltaic cleaning robot and ending the cleaning task.

Compared with the prior art, the present invention has the following advantages: the photovoltaic cleaning robot operation path planning method under complex working conditions introduces a boundary-seeking strategy and a bottom-exploring strategy. The boundary-seeking strategy expands the exploration range of the photovoltaic cleaning robot during operation, while the bottom-exploring strategy ensures that after the photovoltaic cleaning robot enters a new area and completes the cleaning task of the entire area, it can effectively return to the previous area to continue cleaning the unfinished part, thereby improving the efficiency and adaptability of the photovoltaic cleaning robot and enabling it to better cope with the challenges of complex environments.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG1 is a schematic flow chart of a method for planning a path for a photovoltaic cleaning robot under complex working conditions in a first embodiment of the present invention;

FIG2 is a schematic diagram of the positions of the radar module, the IMU module and the diffuse reflection sensor in the photovoltaic cleaning robot in the first embodiment of the present invention; in the figure, 11, left radar; 12, front radar; 13, right radar; 2, diffuse reflection sensor; 3, IMU module;

3 is an example diagram of the running path of the photovoltaic cleaning robot using the boundary-reaching line-changing strategy in the second embodiment of the present invention (top view);

4 is an example diagram of the running path of the photovoltaic cleaning robot using the new area exploration strategy in the third embodiment of the present invention (top view);

5 is an example diagram (top view) of the operation path of the photovoltaic cleaning robot in the fourth embodiment of the present invention using the photovoltaic cleaning robot operation path planning method under complex working conditions of the present invention.

Detailed ways

The following is a detailed description of an embodiment of the present invention. This embodiment is implemented on the premise of the technical solution of the present invention, and a detailed implementation method and a specific operation process are given, but the protection scope of the present invention is not limited to the following embodiment.

Embodiment 1

This embodiment provides a technical solution: a method for planning the path of a photovoltaic cleaning robot under complex working conditions, based on a simulated random single-channel connection area maze environment, covering a strategy of efficiently completing all paths from the starting point and then returning to the origin, which can be applied to solar photovoltaic cleaning tasks under complex working conditions, including the following main strategies:

State determination strategy: The photovoltaic cleaning robot determines the boundary of the initial position as upward and sets the initial forward direction to the right.

During the operation of the photovoltaic cleaning robot, whether the current position is a left corner or a right corner is determined based on the relationship between the running direction, the detection distance values of the front radar, the left radar, and the right radar and the set distance value. For example: when the photovoltaic cleaning robot moves downward, when the detection distance values of the front radar and the left radar (the distance values obtained by detection) exceed the set distance value, and the detection distance value of the right radar is within the set distance value, the current position is regarded as a right corner (see mark ② in Figure 5), and when the photovoltaic cleaning robot moves to the left, when the detection distance values of the front radar and the left radar exceed the set distance value, and the detection distance value of the right radar is within the set distance value, the current position is regarded as a left corner (see mark ⑤ in Figure 5).

It should be noted that the left corner is the junction of the lower boundary and the left boundary of the area above the lower boundary, and the right corner is the junction of the lower boundary and the right boundary of the area above the lower boundary; wherein, taking the initial position as the reference, when the bottom of any photovoltaic panel is not connected to other photovoltaic panels, the bottom of the photovoltaic panel is the lower boundary or part of the lower boundary.

It should be noted that when the driving direction is different, the judgment conditions for left turn or right turn are also different, but they are all judged based on the driving direction and the relationship between the detection distance values of the front radar, left radar, and right radar and the set distance value.

Moreover, the positions under different driving directions trigger different execution steps, as shown in Table 1. This means that according to the position of the photovoltaic cleaning robot and the feedback information from the sensor, the robot will take different actions to adapt to different situations to achieve efficient cleaning tasks.

Table 1 Different execution steps triggered by positions in different driving directions

It should be noted that, in the above table, "normal" means that the detection distance value of the corresponding radar is within the set distance value, and "out of bounds" means that the detection distance value of the corresponding radar exceeds the set distance value. All radars are ultrasonic radars. The specific position of the photovoltaic cleaning robot is shown in Figure 2. As shown in Figure 2, the left radar 11, the front radar 12, and the right radar 13 are all installed at the lower end of the photovoltaic cleaning robot, and all detect in the direction perpendicular to the plane where the photovoltaic panels are located. Among them, the front radar 12 is located in front of the left radar 11 and the right radar 13, and the left radar 11 and the right radar 13 are located on the same straight line perpendicular to the axis of the photovoltaic cleaning robot, and the straight line is parallel to the plane where the photovoltaic cleaning robot is located (the surface of the photovoltaic panel to be cleaned).

It should be noted that the photovoltaic cleaning robot in this embodiment includes the following main monitoring modules (see Figure 2):

IMU module 3: installed inside the photovoltaic cleaning robot, used to measure the acceleration and angular velocity of the photovoltaic cleaning robot to obtain the accurate driving direction.

Radar module: includes a left radar 11, a front radar 12, and a right radar 13, and is used to detect the distance value from the photovoltaic panel by transmitting and receiving radio waves.

Diffuse reflection sensor 2: There are 4 sensors installed at the lower end of the photovoltaic cleaning robot to identify the white grid lines on the photovoltaic panels to improve the accuracy of the robot's running direction and positioning ability, thereby assisting it in driving in a straight line without deviation.

It should be noted that, by collecting characteristic information at the corresponding positions of the photovoltaic panels below, a plurality of diffuse reflection sensors 2 are used to assist the photovoltaic cleaning robot to avoid deflection when driving in a straight line; when driving in a normal straight line, the two diffuse reflection sensors 2 arranged along the axial direction are located directly above the white grid lines on the photovoltaic panels, and the other two diffuse reflection sensors 2 are located on both sides of the white grid lines. The diffuse reflection sensors 2 at different positions collect different characteristic information, and the two diffuse reflection sensors 2 arranged along the axial direction are controlled to be located directly above the white grid lines on the photovoltaic panels, thereby assisting the photovoltaic cleaning robot to avoid deflection when driving in a straight line.

Line-changing strategy: This strategy stipulates that when the PV cleaning robot reaches the left and right boundaries of the PV panel, the robot should perform a downward line-changing action and then run in the opposite direction to continue the cleaning task. The implementation of this strategy helps ensure that the robot can efficiently complete the comprehensive cleaning of the PV panel, especially when traveling along the boundary to avoid uncleaned areas.

Boundary-seeking strategy: Boundary-seeking strategy is a key strategy used to optimize the operation logic of the photovoltaic cleaning robot. During straight-line driving, this strategy relies on the detection of the left radar to identify new cleaning areas. The premise is to follow the bottom-finding strategy. Once the left radar detects a new cleaning area, the photovoltaic cleaning robot will run to the new area to start the cleaning task.

Bottom-probing strategy: Its main purpose is to ensure that the photovoltaic cleaning robot will return to the previous area only after completely cleaning the current area, avoiding returning to the bottom of the original area in advance, resulting in uncleaned parts in the current area. This strategy is based on the following principles:

When the photovoltaic cleaning robot is cleaning in the current area, if it reaches the right corner or the left corner, it will get a chance to explore the bottom. If it is driving straight to the left at this time, when the detection distance value of the left radar changes from exceeding the set distance value to within the set distance value, it indicates that a new area has appeared below the front. Since the photovoltaic cleaning robot has reached the right corner before and got a chance to explore the bottom, it turns left to enter the new area below the front (for example, the downward exploration action at the mark ③ in Figure 5). After running downward once, the opportunity to explore the bottom will be cancelled. When a new area appears below the front again, the photovoltaic cleaning robot performs a U-turn (see the U-turn on the right side of the mark ④ in Figure 5) and returns. This behavior ensures that the photovoltaic cleaning robot avoids entering the new lower area in advance without the opportunity to explore the bottom, so as to continue working in the cleaned area.

The triggering method of the bottom exploration opportunity is as follows: once the photovoltaic cleaning robot reaches the left or right corner during operation, it will get a bottom exploration opportunity. During the process of the photovoltaic cleaning robot running to the left, if the detection distance value of the left radar changes from within the set distance value to beyond the set distance value, the photovoltaic cleaning robot will continue to move forward. During the process of running to the left, if the detection distance value of the left radar changes from beyond the set distance value to within the set distance value, it indicates that there is a lower area ahead that needs to be explored downward; at this time, the photovoltaic cleaning robot will turn left, enter the lower area, and execute the line change strategy to move downward line by line.

It is important to note that once the PV cleaning robot completes a downward run (including the downward run in the line-changing strategy, but excluding the U-turn operation), the bottom exploration opportunity will be cancelled, so the PV cleaning robot can only get one bottom exploration opportunity at most. This strategy ensures that the PV cleaning robot performs the bottom exploration action at the appropriate time to effectively explore the area below, while avoiding repeated bottom exploration behavior.

The following is a further explanation of the working process of the photovoltaic cleaning robot:

As shown in Figure 1, according to the logic of the boundary-finding strategy and the bottom-probing strategy, the boundary position of an area is taken as the starting point (initial position). The photovoltaic cleaning robot determines the boundary of the initial position as the top, sets the initial forward direction to the right, and the detection distance value of the left radar of the photovoltaic cleaning robot at the initial position exceeds the set distance value, and the detection distance values of the front radar and the right radar are within the set distance value, ensuring that when the photovoltaic cleaning robot runs along the boundary, the left detection radar is outside the boundary, which is convenient for detecting new areas. At the beginning of cleaning, the robot adopts the line-breaking strategy and advances downward line by line. When the photovoltaic cleaning robot is running, the robot constantly monitors the feedback of the left radar. Once the detection distance value of the left radar decreases from exceeding the set distance value to within the set distance value, it indicates that a new area appears on the left, and the boundary-finding strategy is executed. The robot will turn left and enter the new area to continue running. During the running process, the line-breaking strategy and the boundary-finding strategy are continued, and the bottom-probing strategy is followed until the cleaning task is terminated after reaching the initial position.

It should be noted that in this embodiment, the initial position area is set to all white or all black, so that when the photovoltaic cleaning robot reaches the initial position area, the characteristic information is recognized by multiple diffuse reflection sensors 2, and the current position can be confirmed as the initial position, thereby terminating the operation of the photovoltaic cleaning robot and ending the cleaning task.

When the robot reaches the left or right corner, it will get one chance to explore the bottom. If the robot moves downward once (including downward movement when changing lines, but excluding the situation when turning around), the chance to explore the bottom will be cancelled. There is only one chance to explore the bottom at most.

When the robot reaches the left corner while running to the left, if the bottom-finding strategy is met, the robot will start to run clockwise along the boundary (without the switching action) until it returns to the initial position. This strategy ensures that the photovoltaic cleaning robot completes the comprehensive cleaning task without missing any area.

In this embodiment, the directions of "up", "down", "left" and "right" during the operation of the photovoltaic cleaning robot are consistent with the directions of "up", "down", "left" and "right" in the plan view of the cleaning task execution area.

Embodiment 2

According to the situation shown in Figure 3, when the photovoltaic cleaning robot adopts the strategy of reaching the boundary to change lines, two uncleaned areas appear. At mark ①, the photovoltaic cleaning robot fails to explore the new area to the upper right of mark ①, while at mark ②, the area to the lower right of mark ② is not cleaned, but returns to the initial area in advance. This shows that the photovoltaic cleaning robot needs to improve its path planning strategy to ensure more comprehensive coverage and cleaning of all areas.

In this embodiment, the directions of "up", "down", "left" and "right" during the operation of the photovoltaic cleaning robot are consistent with the directions of "up", "down", "left" and "right" in the plan view of the cleaning task execution area.

Embodiment 3

According to the situation shown in Figure 4, when the photovoltaic cleaning robot uses the new area exploration strategy, there are two uncleaned areas, namely the lower right area and the lower left area. At mark ①, since the photovoltaic cleaning robot will explore the new area, it goes to the top of the new area. However, at marks ② and ③, the photovoltaic cleaning robot fails to fully clean the area below the current area before returning to the previous area, resulting in the appearance of missed areas. To ensure comprehensive cleaning, the strategy needs to be improved to avoid this situation.

In this embodiment, the directions of "up", "down", "left" and "right" during the operation of the photovoltaic cleaning robot are consistent with the directions of "up", "down", "left" and "right" in the plan view of the cleaning task execution area.

Embodiment 4

According to the situation shown in FIG5 , when the photovoltaic cleaning robot is operated using the photovoltaic cleaning robot operation path planning method under complex working conditions of the present invention, all areas are successfully cleaned. At mark ①, the photovoltaic cleaning robot timely explores the upper area and executes the line-breaking strategy after reaching the upper boundary to ensure cleaning. At mark ②, it reaches the right corner, triggering a bottoming opportunity. However, at mark ③, the robot exhausts a bottoming opportunity when moving downward, so it turns around and returns at mark ④.

When the robot reaches the mark ⑤, it moves to the left and reaches the left corner, and moves upward to reach the mark ④ again. At this time, it has a chance to explore the bottom, continue forward, and return to the initial area to ensure that the original area is completely cleaned before returning to the previous area. This strategy ensures the effective operation and comprehensive cleaning of the photovoltaic cleaning robot during the cleaning process.

In this embodiment, the directions of "up", "down", "left" and "right" during the operation of the photovoltaic cleaning robot are consistent with the directions of "up", "down", "left" and "right" in the plan view of the cleaning task execution area.

It should be noted that each block in Figure 5 is a photovoltaic panel, and the two areas on the left and right sides are connected by a photovoltaic panel, that is, the two areas are connected by a single channel, and the width of the channel is one photovoltaic panel (when the number of channels is greater than or equal to two, the method of the present invention cannot be applied. In actual situations, photovoltaic panels between multiple areas are generally connected by a single channel). The present invention is only applicable to complex working conditions formed by multiple areas connected by a single channel.

To sum up, the photovoltaic cleaning robot operation path planning method under complex working conditions in the above-mentioned embodiment introduces a boundary-seeking strategy and a bottom-exploring strategy. The boundary-seeking strategy expands the exploration range of the photovoltaic cleaning robot during operation, while the bottom-exploring strategy ensures that after entering a new area and completing the cleaning task of the entire area, the photovoltaic cleaning robot can effectively return to the previous area to continue cleaning the unfinished parts, thereby improving the efficiency and adaptability of the photovoltaic cleaning robot and enabling it to better cope with the challenges of complex environments.

Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations of the present invention. A person skilled in the art may change, modify, replace and vary the above embodiments within the scope of the present invention.

Claims (2)

1. The method for planning the running path of the photovoltaic cleaning robot under the complex working condition is characterized by comprising the following steps of:

S1: at any boundary position of the cleaning task execution area as an initial position, the photovoltaic cleaning robot determines the boundary of the initial position as the upper part, and sets the initial advancing direction as the right side;

S2: after cleaning is started, the photovoltaic cleaning robot executes a line feed strategy and advances line by line downwards; when the photovoltaic cleaning robot runs, the photovoltaic cleaning robot continuously monitors feedback of the left radar, acquires a detection distance value of the left radar in real time, and when the detection distance value of the left radar is reduced from exceeding a set distance value to within the set distance value, a new area appears on the left side, a boundary searching strategy is executed, and the robot turns left and enters the new area;

S3: the photovoltaic cleaning robot turns left and enters a new area and then continues to run, a line changing strategy and a boundary searching strategy are continuously executed in the running process, a bottom searching strategy is followed, and the cleaning task is ended until the photovoltaic cleaning robot returns to an initial position;

In the steps S1 to S3, when the photovoltaic cleaning robot runs along the boundary, the left radar position of the photovoltaic cleaning robot is positioned outside the boundary, and the detection distance value exceeds a set value;

The left radar, the front radar and the right radar are all arranged at the lower end of the photovoltaic cleaning robot and are all detected in the direction perpendicular to the plane where the photovoltaic panel is located, wherein the front radar is positioned in front of the left radar and the right radar, the left radar and the right radar are positioned on the same straight line perpendicular to the axis of the photovoltaic cleaning robot, and the straight line is parallel to the plane where the photovoltaic cleaning robot is located;

In the steps S2 and S3, during the operation of the photovoltaic cleaning robot, whether the current position is a left corner or a right corner is determined according to the operation direction and the relation between the detection distance values of the front radar, the left radar and the right radar and the set distance value;

in the step S2, the line feed strategy is as follows: when the photovoltaic cleaning robot reaches the left and right boundaries of the photovoltaic cell panel, the robot performs downward line feed action and then runs in the opposite direction so as to continue the cleaning task;

in the step S3, when the boundary finding policy is executed, a bottom finding policy needs to be followed, where the bottom finding policy is as follows: when the photovoltaic cleaning robot cleans in the current area, if the right corner or the left corner is not reached, the bottom detection opportunity is not obtained, and the left straight running is being executed at the moment, when the detection distance value of the left radar is changed from exceeding the set distance value to being within the set distance value, a new area appears below the front, and the photovoltaic cleaning robot cannot enter the new area below the front in advance because the photovoltaic cleaning robot does not reach the left corner and the right corner at the bottom end of the current area, and the photovoltaic cleaning robot executes the turning operation and returns;

The triggering mode of the bottom detection opportunity is as follows: when the photovoltaic cleaning robot reaches a left corner or a right corner in operation, a bottom detection opportunity is obtained, and in the process of the photovoltaic cleaning robot running leftwards, if the detection distance value of the left radar is changed from within a set distance value to exceed the set distance value, the photovoltaic cleaning robot continues to advance; in the process of running to the left, if the detection distance value of the left radar is changed from exceeding the set distance value to being within the set distance value, the fact that a lower area exists in front of the left radar is needed to be searched downwards is indicated; at the moment, the photovoltaic cleaning robot with the bottom detection opportunity executes left turning action, enters a lower area, executes a line changing strategy and advances downwards line by line;

The left corner is the junction of the lower boundary and the left boundary of the area above the lower boundary, and the right corner is the junction of the lower boundary and the right boundary of the area above the lower boundary; when the lower part of any photovoltaic cell panel is not connected with other photovoltaic cell panels by taking the initial position as a reference, the lower edge of the photovoltaic cell panel is the lower boundary or a part of the lower boundary;

When the photovoltaic cleaning robot completes one downward running action, the bottom detection opportunity is canceled, and the downward running action comprises the downward running action in the line changing strategy, but does not comprise the turning-around action.

2. The method for planning the running path of the photovoltaic cleaning robot under the complex working condition according to claim 1 is characterized in that: in the step S3, the initial position area is set to be full white or full black, and when the photovoltaic cleaning robot reaches the initial position area, the characteristic information of the initial position area is identified through the plurality of diffuse reflection sensors arranged at the lower end of the photovoltaic cleaning robot, so that the current position can be confirmed to be the initial position, the operation of the photovoltaic cleaning robot is further finished, and the cleaning task is finished.