Substation inspection robot autonomous inspection navigation method based on laser radar

Abstract

The invention discloses a transformer substation inspection robot autonomous inspection navigation method based on a laser radar, and relates to the technical field of transformer substation inspection, and the method comprises the steps: collecting transformer substation environment information through the laser radar, obtaining a laser data point set, and constructing a two-dimensional grid map based on the laser data point set; planning an optimal path by using the two-dimensional grid map after a target location is received; the optimal path is defined by integrating distance, safety, equipment importance and obstacle density; and carrying out routing inspection according to the optimal path, carrying out real-time environment perception and self-positioning, and judging whether to enter an obstacle avoidance mode so as to realize autonomous routing inspection. According to the method, the 360-degree laser radar is utilized to construct the high-precision three-dimensional point cloud map, and the high-precision three-dimensional point cloud map is converted into the two-dimensional grid map, so that abundant obstacle information is reserved, and the calculation of path planning is simplified. An improved Dijkstra algorithm is adopted, safety, equipment importance and obstacle density are taken into consideration of path planning, and an optimal inspection path is generated.

Description

Technical Field

[0001] The present invention relates to the technical field of substation inspection, and in particular to an autonomous inspection navigation method of a substation inspection robot based on laser radar. Background Art

[0002] In recent years, substation inspection technology has been widely used in power system maintenance. Traditional substation inspection mainly relies on manual inspection, which is not only time-consuming and labor-intensive, but also has certain safety hazards. With the development of science and technology, automated inspection technology has gradually emerged, especially the inspection method based on robot technology has received more and more attention. Existing substation inspection robots mainly use a variety of sensing methods such as visual sensors and ultrasonic sensors to conduct inspections through preset routes. However, these methods have limited perception and navigation capabilities in complex environments and are easily affected by factors such as environmental changes and obstacle interference, resulting in low inspection efficiency and accuracy. As a high-precision environmental perception device, LiDAR can provide accurate distance information and environmental scanning data, and has demonstrated its unique advantages in the fields of autonomous driving and drone navigation.

[0003] Although LiDAR-based navigation technology has been applied in many fields, it still faces many challenges in the field of substation inspection. First, the internal environment of the substation is complex and there are many devices. Traditional LiDAR navigation methods are difficult to plan efficient and safe inspection paths in such a high-density obstacle environment. Secondly, the existing technology has deficiencies in real-time environmental perception and obstacle avoidance, and cannot ensure that the inspection robot always maintains accurate navigation in a dynamically changing environment. In addition, most of the existing path planning methods aim at the shortest path, ignoring factors such as the importance of equipment and obstacle density, resulting in higher risks during the inspection process. Therefore, how to achieve efficient and safe autonomous inspection navigation in the complex environment of the substation has become the focus and difficulty of current technological development. Summary of the invention

[0004] In view of the problems existing in the existing inspection methods, the present invention is proposed.

[0005] Therefore, the problem to be solved by the present invention is that the internal environment of the power station is complex and there are many devices. The traditional lidar navigation method is difficult to plan an efficient and safe inspection path in such an environment with high-density obstacles, and obstacle avoidance cannot be effectively implemented.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, an embodiment of the present invention provides a laser radar-based substation inspection robot autonomous inspection navigation method, which includes:

[0008] Collecting substation environmental information by laser radar to obtain a laser data point set, and constructing a two-dimensional grid map based on the laser data point set;

[0009] After receiving the target location, the optimal path is planned using the two-dimensional grid map; the optimal path is defined based on the comprehensive distance, safety, equipment importance and obstacle density;

[0010] According to the optimal path inspection, real-time environmental perception and self-positioning are performed to determine whether to enter the obstacle avoidance mode to achieve autonomous inspection.

[0011] As a preferred solution of the autonomous inspection navigation method of the substation inspection robot based on laser radar of the present invention, wherein: the construction of the two-dimensional grid map is based on the three-dimensional point cloud map constructed by the laser data point set and converted into a two-dimensional grid map while retaining obstacle information;

[0012] The conversion of 3D point cloud map to 2D raster map includes:

[0013] A height threshold suitable for the substation environment is set, the entire three-dimensional point cloud data is traversed, and all points below the threshold are projected onto a two-dimensional plane. Then, the two-dimensional plane is divided into uniform grids, and each grid represents a unit in the grid map. Subsequently, the number of points in each grid is counted. If the number of points exceeds the preset threshold, the grid is marked as occupied, otherwise it is marked as free. Finally, the generated two-dimensional grid map is smoothed to remove isolated grids caused by noise.

[0014] As a preferred solution of the autonomous inspection navigation method of the substation inspection robot based on laser radar of the present invention, the optimal path is planned based on the comprehensive distance, safety, equipment importance and obstacle density, as shown in the following formula:

[0015]

[0016] Among them, P opt is the optimal path, p is a possible path in the path set P, n is the number of nodes on the path, C( vi ,v i+1 ) is a comprehensive index combining various factors, v i and v i+1 is a node;

[0017] The C( vi ,v i+1 ) is shown in the following formula:

[0018] C(v i ,v i+1 )=C e (v i,v i+1 )·(1+γ·O(v i ,v i+1 ))

[0019] Among them, O( vi ,v i+1 ) is the obstacle density function between nodes, γ is the obstacle influence weight, C e ( vi ,v i+1 ) To ensure that important equipment indicators are met;

[0020] The C e ( vi ,v i+1 ) is shown in the following formula:

[0021] C e (v i ,v i+1 )=C s (v i ,v i+1 )·(1-β·E(v i ,v i+1 ))

[0022] Among them, E( vi ,v i+1 ) is the device importance function between nodes, β is the importance weight, C s ( vi ,v i+1 ) To ensure safety indicators;

[0023] The C s ( vi ,v i+1 ) is shown in the following formula:

[0024] C s (v i ,v i+1 )=C d (v i ,v i+1 )·(1+a·S(v i ,v i+1 ))

[0025] Among them, S( vi ,v i+1 ) is the security level function between nodes, α is the security level weight, C d ( vi ,v i+1 ) is the basic distance cost indicator to ensure the shortest path;

[0026] The C d ( vi ,vi+1 ) is shown in the following formula:

[0027] C d (v i ,v i+1 )=D(v i ,v i+1 )

[0028] Among them, D( vi ,v i+1 ) is the node v i and v i+1 The Euclidean distance between

[0029] The O( vi ,v i+1 )、E( vi ,v i+1 )、S( vi ,v i+1 ) and D( vi ,v i+1 ) is shown in the following formula:

[0030]

[0031] S(v i ,v i+1 )=max(S i ,S i+1 ) / S max

[0032] E(v i ,v i+1 )=(E i +E i+1 ) / (2·E max )'

[0033] O(v i ,v i+1 ) = count(obstacles) / max obstacles

[0034] Among them, x i and i For node v i The coordinates of S i For node v i The security level, S max The highest security level, E i For node v i The importance value of nearby devices, E max is the highest importance value, count(obstacles) is the number of obstacles on the path segment, max obstacles The maximum number of obstacles that can be set.

[0035] As a preferred solution of the laser radar-based substation inspection robot autonomous inspection navigation method of the present invention, the determination of whether to enter the obstacle avoidance mode includes:

[0036] When the robot is inspecting in a substation environment, the system continuously uses lidar to scan the surrounding 360-degree environment, focusing on analyzing the data in front, on the left, and on the right, and determining the safety factor in front. At the same time, the system uses lidar data and point cloud processing algorithms to estimate the size of detected obstacles, compares the safety factor with a preset first threshold, and compares the estimated obstacle size with a preset second threshold. When the safety factor is lower than the first threshold or the obstacle size is greater than the second threshold, it further evaluates whether the obstacle interferes with the robot's normal travel path. If so, it enters an obstacle avoidance state, otherwise it enters a non-obstacle avoidance state.

[0037] As a preferred solution of the autonomous inspection navigation method of the substation inspection robot based on laser radar of the present invention, wherein: in the obstacle avoidance state, multiple candidate offset points are generated around the optimal path, and the best offset waypoint is selected by comprehensively evaluating the safety of each candidate point, the degree of deviation from the original path and the impact on the subsequent path, and after the offset waypoint is determined, a local path is quickly generated between the original global path and the new offset point;

[0038] The determination of the offset waypoint is shown in the following formula:

[0039]

[0040] Among them, P offset is the best offset waypoint; C is the set of candidate offset points; α, β, γ are weight coefficients, and α+β+γ=1; F total (p) is the improved artificial potential field function; d(p,P original ) is the candidate point p and the original path point P original distance; I(p) is the influence factor of candidate point p on the subsequent path.

[0041] As a preferred solution of the autonomous inspection navigation method of the substation inspection robot based on laser radar described in the present invention, wherein: in the non-obstacle avoidance state, it is determined whether the obstacle is a dynamic obstacle. If so, the relative motion state between the obstacle in front and the robot is predicted based on the Kalman filter, and the speed of the robot is adjusted in real time through the adaptive cruise control algorithm in combination with the speed constraint conditions unique to the substation;

[0042] If a static obstacle is detected, the system first evaluates the position, size and relationship of the obstacle to the robot's planned path. If the static obstacle is judged to interfere with the robot's normal path and cannot be circumvented by changing the steering angle, it switches to obstacle avoidance mode. If the static obstacle can be circumvented by changing the steering angle, two steering angles are determined to bypass the obstacle.

[0043] As a preferred solution of the autonomous inspection navigation method of the substation inspection robot based on laser radar of the present invention, the two-stage steering angle is determined according to the obstacle area, channel width and distance, as shown in the following formula:

[0044]

[0045] Among them, θ base is the basic obstacle avoidance angle, w obs is the obstacle width, d safe is the safety distance, d obs is the distance from the obstacle to the robot, A obs is the obstacle area, A max is the preset maximum obstacle area, w channel is the channel width, R max is the maximum impact distance.

[0046] In a second aspect, an embodiment of the present invention provides a substation inspection robot autonomous inspection navigation system based on laser radar, which includes:

[0047] A data acquisition module is used to collect substation environmental information through a laser radar to obtain a laser data point set, and to construct a two-dimensional grid map based on the laser data point set;

[0048] A path planning module, used to plan an optimal path using the two-dimensional grid map after receiving the target location; the optimal path is defined by comprehensive distance, safety, equipment importance and obstacle density;

[0049] The navigation control module is used to inspect according to the optimal path, perform real-time environmental perception and self-positioning, determine whether to enter the obstacle avoidance mode, and realize autonomous inspection.

[0050] In a third aspect, an embodiment of the present invention provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the processor executes the computer program, it implements any step of the above-mentioned lidar-based substation inspection robot autonomous inspection and navigation method.

[0051] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, any step of the above-mentioned lidar-based substation inspection robot autonomous inspection and navigation method is implemented.

[0052] The beneficial effect of the present invention is to use 360-degree laser radar and SLAM algorithm to construct a high-precision three-dimensional point cloud map, and convert it into a two-dimensional grid map, which not only retains rich obstacle information, but also simplifies the calculation of path planning. The improved Dijkstra algorithm is used to take safety, equipment importance and obstacle density into consideration in path planning to generate the optimal inspection path. Through real-time environmental perception and pose estimation, combined with multi-level safety threshold functions and particle filtering algorithms, it is ensured that the robot can accurately avoid obstacles and adjust the path during the inspection process, and realize adaptive inspection in dynamic environments. The organic combination of these methods not only improves the efficiency and safety of inspections, but also significantly reduces the computational complexity, demonstrating the innovative ability of efficient autonomous navigation in complex environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. Among them:

[0054] Figure 1 This is a flow chart of the autonomous inspection and navigation method of a substation inspection robot based on lidar.

[0055] Figure 2 The robot structure diagram of the autonomous inspection navigation method of the substation inspection robot based on lidar.

[0056] Figure 3 This is a structural diagram of the navigation system of the autonomous inspection navigation method of the substation inspection robot based on lidar. DETAILED DESCRIPTION

[0057] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0058] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0059] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0060] The present invention is described in detail with reference to schematic diagrams. When describing the embodiments of the present invention, for the sake of convenience, the cross-sectional diagrams showing the device structure will not be partially enlarged according to the general scale, and the schematic diagrams are only examples, which should not limit the scope of protection of the present invention. In addition, in actual production, the three-dimensional dimensions of length, width and depth should be included.

[0061] At the same time, in the description of the present invention, it should be noted that the directions or positional relationships indicated by the terms "upper, lower, inner and outer" are based on the directions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore cannot be understood as limiting the present invention. In addition, the terms "first, second or third" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0062] In the present invention, unless otherwise clearly specified and limited, the terms "install, connect, connect" should be understood in a broad sense, for example: it can be a fixed connection, a detachable connection or an integral connection; it can also be a mechanical connection, an electrical connection or a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0063] Example 1

[0064] Reference Figure 1 and Figure 2 , which is the first embodiment of the present invention, and provides a laser radar-based substation inspection robot autonomous inspection navigation method, comprising:

[0065] S1: Collecting substation environmental information through laser radar to obtain a laser data point set, and constructing a two-dimensional grid map based on the laser data point set;

[0066] Specifically, the laser radar installed on the top of the inspection robot performs 360-degree unobstructed rotation scanning, and the laser sensor is driven to rotate by controlling the motor to collect environmental information in the substation in all directions. Subsequently, the system uses the approximate simultaneous localization and mapping (SLAM) algorithm to process the collected laser data point set and construct a high-precision three-dimensional point cloud map. This method can not only achieve all-round perception of the robot's surrounding environmental information, but also avoid the blind spots and height restrictions of traditional single-line laser radars.

[0067] Next, the 3D point cloud map is converted into a 2D grid map, which simplifies the subsequent path planning calculations while retaining rich obstacle information.

[0068] Preferably, a height threshold suitable for the substation environment is set, which is usually a value slightly higher than the height of the robot itself; then, the entire three-dimensional point cloud data is traversed, and all points below the threshold are projected onto a two-dimensional plane; then, the two-dimensional plane is divided into uniform grids, each grid representing a unit in the grid map; then, the number of points in each grid is counted, and if the number of points exceeds the preset threshold, the grid is marked as occupied, otherwise it is marked as free; finally, the generated two-dimensional grid map is smoothed to remove isolated grids that may be caused by noise. This method not only retains the contour information of important obstacles and equipment in the substation, but also greatly simplifies the data structure, providing efficient input for subsequent path planning, thereby achieving the goal of significantly reducing computational complexity while retaining key environmental features.

[0069] S2: after receiving the target location, using the two-dimensional grid map to plan the optimal path;

[0070] Specifically, the core of the intelligent navigation system is the path planning module. This module first uses the two-dimensional grid map generated in the early stage to load obstacle information to provide an environmental basis for planning. At the same time, it is closely integrated with the map matching and positioning module to obtain the precise position information of the inspection robot in real time.

[0071] When the system receives the target location instruction, the path planning module is immediately started and the optimal path is quickly generated based on the improved Dijkstra algorithm.

[0072] For example, in a feasible embodiment, the improved Dijkstra algorithm quickly generates the optimal path as follows:

[0073] First, in the original Dijkstra algorithm, the distance between nodes is taken into account, which ensures that the robot chooses the shortest path:

[0074] C d (v i ,v i+1 )=D(vi ,v i+1 )

[0075] Among them, D( vi ,v i+1 ) is the node v i and v i+1 The Euclidean distance between .

[0076] Safety is the primary consideration for substation inspections, which means that during the inspection process, the robot should stay away from large and dangerous equipment as much as possible:

[0077] C s (v i ,v i+1 )=C d (v i ,v i+1 )·(1+a·S(v i ,v i+1 ))

[0078] Among them, S( vi ,v i+1 ) is the security level function between nodes, α is the security level weight, C d ( vi ,v i+1 ) is used as the basic distance cost indicator to ensure the shortest path.

[0079] Furthermore, during the inspection process, the importance of the equipment is given priority, that is, the path is prioritized to pass through the important equipment accessories, thereby further improving the inspection efficiency:

[0080] C e (v i ,v i+1 )=C s (v i ,v i+1 )·(1-β·E(v i ,v i+1 ))

[0081] Among them, E( vi ,v i+1 ) is the device importance function between nodes, β is the importance weight, C s ( vi ,v i+1 ) to ensure safety indicators.

[0082] Finally, according to the obstacle information of the two-dimensional grid map, it is taken into account when designing the path to avoid areas with dense obstacles and improve the safety and smoothness of navigation:

[0083] C(v i ,v i+1 )=Ce (v i ,v i+1 )·(1+γ·O(v i ,v i+1 ))

[0084] Among them, O( vi ,v i+1 ) is the obstacle density function between nodes, γ is the obstacle influence weight, C e ( vi ,v i+1 ) to ensure passing important equipment indicators.

[0085] The formula for the final optimal path is:

[0086]

[0087] Among them, P opt is the optimal path, p is a possible path in the path set P, n is the number of nodes on the path, C( vi ,v i+1 ) is a comprehensive index combining various factors, v i and v i+1 For the node.

[0088] It should be noted that the functions involved in the optimal path are defined as follows:

[0089]

[0090] S(v i ,v i+1 )=max(S i ,S i+1 ) / S max

[0091] E(v i ,v i+1 )=(E i +E i+1 ) / (2·E max )'

[0092] O(v i ,v i+1 ) = count(obstacles) / max obstacles

[0093] Among them, x i and i For node v i The coordinates of S i For node v i The security level, S max The highest security level, E iFor node v i The importance value of nearby devices, E max is the highest importance value, count(obstacles) is the number of obstacles on the path segment, max obstacles The maximum number of obstacles that can be set.

[0094] After planning is completed, the module will generate a series of precise motion control instructions and transmit them directly to the robot's underlying drive system. This seamless integration from high-level planning to low-level control ensures that the inspection robot can move smoothly and safely along the calculated optimal path.

[0095] S3: Inspect according to the optimal path, perform real-time environmental perception and self-positioning, determine whether to enter obstacle avoidance mode, and implement autonomous inspection.

[0096] Furthermore, in order to improve navigation accuracy, a navigation control module was developed to ensure that the robot can accurately reach the specified location through target position correction and path tracking algorithms. This module first uses the real-time scanning data of the lidar to match the pre-built environment map, and continuously optimizes the robot's pose estimation through the particle filter algorithm.

[0097] When the robot is inspecting in the substation environment, the system continuously uses the lidar to scan the surrounding 360-degree environment at a high frequency, focusing on analyzing the data in front, on the left, and on the right. Through the designed multi-level safety threshold function, the system calculates the safety factor in front in real time, which comprehensively considers the distance of obstacles, relative speed, and the danger level of substation equipment. At the same time, the system uses lidar data and point cloud processing algorithms to estimate the size of detected obstacles. The system compares the calculated safety factor with the preset safety threshold, and compares the estimated obstacle size with the predefined size threshold (set according to the characteristics of substation equipment and channel width). When the safety factor is lower than the preset threshold, or the obstacle size exceeds the predefined threshold, the system will further evaluate the position, shape, and relationship of the obstacle to the robot's current path. If the obstacle is judged to be likely to interfere with the robot's normal path, the system will immediately trigger the obstacle avoidance state.

[0098] In the obstacle avoidance state, a series of candidate offset points are first generated around the optimal path using a fast scanning algorithm. Then, the best offset waypoint is selected by comprehensively evaluating the safety of each candidate point, the degree of deviation from the original path, and the impact on the subsequent path. After determining the offset waypoint, the system immediately starts the local path replanning module to quickly generate a smooth and safe local path between the original global path and the new offset point.

[0099] It should be noted that the best offset waypoint uses an improved artificial potential field method combined with a Bezier curve interpolation algorithm to quickly generate smooth offset waypoints, as shown in the following formula:

[0100]

[0101] Among them, P offset is the best offset waypoint; C is the set of candidate offset points; α, β, γ are weight coefficients, and α+β+γ=1; F total (p) is the improved artificial potential field function; d(p,P original ) is the candidate point p and the original path point P original distance; I(p) is the influence factor of candidate point p on the subsequent path.

[0102] Among them, the improved artificial potential field function F total (p) is shown in the following formula:

[0103]

[0104] Among them, k att is the gravitational constant; k rep is the repulsive force constant; P goal is the target point; i is the i-th obstacle; d 0 is the influence range of the obstacle; λ is the attenuation coefficient.

[0105] The impact factor on the subsequent path is defined as follows:

[0106]

[0107] Wherein, B(t) is a second-order Bezier curve with the current point, the offset point p and the next path point as control points.

[0108] Finally, we use the Bezier curve interpolation algorithm to generate a smooth path. At the same time, the navigation control module will adjust the robot's motion parameters, including speed, acceleration, and steering angle, to ensure that the robot can smoothly transition to the new local path. In addition, the system will continue to monitor environmental changes. Once the original obstacle disappears or a new obstacle appears, the path evaluation and adjustment process will be triggered again, thereby achieving real-time adaptation to the dynamic environment.

[0109] If the system is in a non-obstacle avoidance state, it determines whether the obstacle is a dynamic obstacle. If so, the forward collision time (TTC) estimation module is started. This module predicts the relative motion state of the obstacle in front and the robot based on the Kalman filter, and combines the speed constraints unique to the substation to adjust the robot speed in real time through the adaptive cruise control algorithm.

[0110] If a static obstacle is detected, the system first evaluates the position, size and relationship of the obstacle to the robot's planned path. If the static obstacle is judged to be likely to interfere with the robot's normal path and cannot be circumvented by changing the steering angle, it switches to obstacle avoidance mode. If the static obstacle can be circumvented by changing the steering angle, two steering angles are determined to bypass the obstacle.

[0111] Specifically, the two-stage steering angles are determined as follows:

[0112]

[0113] Among them, θ base is the basic obstacle avoidance angle, w obs is the obstacle width, d safe is the safety distance, d obs is the distance from the obstacle to the robot, A obs is the obstacle area, A max is the preset maximum obstacle area, w channel is the channel width, R max is the maximum impact distance.

[0114] Example 2

[0115] This is the second embodiment of the present invention, which provides a laser radar-based substation inspection robot autonomous inspection navigation system, including.

[0116] A data acquisition module is used to collect substation environmental information through a laser radar to obtain a laser data point set, and to construct a two-dimensional grid map based on the laser data point set;

[0117] A path planning module, used to plan an optimal path using the two-dimensional grid map after receiving the target location; the optimal path is defined by comprehensive distance, safety, equipment importance and obstacle density;

[0118] The navigation control module is used to inspect according to the optimal path, perform real-time environmental perception and self-positioning, determine whether to enter the obstacle avoidance mode, and realize autonomous inspection.

[0119] The data acquisition module transmits the collected data to the navigation system, and the navigation system and the inspection robot realize two-way data transmission, the navigation system and the big database realize two-way data transmission, and the navigation system and the control terminal realize two-way data transmission;

[0120] The inspection robot includes a central processor, a laser radar, a drive module, a power supply module, an encoder, a wireless communication module and an anti-collision module;

[0121] The navigation system includes a processing system, a path planning module, a navigation control module and an inspection task receiving module;

[0122] The input end of the central processing unit is connected to the output end of the laser radar, the driving module, the power supply module, the encoder, and the anti-collision module, and the central processing unit is bidirectionally connected to the wireless communication module.

[0123] The laser radar is used to laser scan the indoor environment to obtain a laser data point set. The anti-collision module is used to measure the distance according to the obstacle information scanned by the laser radar, and control the inspection robot to avoid obstacles through a program algorithm. The central processor is used to construct a two-dimensional grid map of the environment in which the inspection robot is located and the optimal path for directional navigation according to the laser data point set. The driving module includes a driving wheel and a driven wheel arranged at the bottom of the inspection robot.

[0124] The path planning module uses the grid map to load obstacle information, and uses the inspection robot's posture output by the map matching and positioning module. After receiving the target location, it starts path planning and outputs motion control instructions to the bottom-level driver of the inspection robot to realize the navigation function.

[0125] The navigation control module continuously uses the lidar to scan the surrounding 360-degree environment at a high frequency, focusing on analyzing the data in front, on the left, and on the right. Through the designed multi-level safety threshold function, the system calculates the safety factor in front in real time, which comprehensively considers the distance to the obstacle, the relative speed, and the danger level of the substation equipment. At the same time, the system uses the lidar data and the point cloud processing algorithm to estimate the size of the detected obstacle. The system compares the calculated safety factor with the preset safety threshold, and compares the estimated obstacle size with the predefined size threshold (set according to the characteristics of the substation equipment and the width of the channel). When the safety factor is lower than the preset threshold, or the obstacle size exceeds the predefined threshold, the system will further evaluate the position, shape, and relationship of the obstacle to the current path of the robot. If the obstacle is judged to be likely to interfere with the normal path of the robot, the system will immediately trigger the obstacle avoidance state.

[0126] In the obstacle avoidance state, a series of candidate offset points are first generated around the optimal path using a fast scanning algorithm. Then, the best offset waypoint is selected by comprehensively evaluating the safety of each candidate point, the degree of deviation from the original path, and the impact on the subsequent path. After determining the offset waypoint, the system immediately starts the local path replanning module to quickly generate a smooth and safe local path between the original global path and the new offset point.

[0127] If the system is in a non-obstacle avoidance state, it determines whether the obstacle is a dynamic obstacle. If so, the forward collision time (TTC) estimation module is started. This module predicts the relative motion state of the obstacle in front and the robot based on the Kalman filter, and combines the speed constraints unique to the substation to adjust the robot speed in real time through the adaptive cruise control algorithm.

[0128] If a static obstacle is detected, the system first evaluates the position, size and relationship of the obstacle to the robot's planned path. If the static obstacle is judged to be likely to interfere with the robot's normal path and cannot be circumvented by changing the steering angle, it switches to obstacle avoidance mode. If the static obstacle can be circumvented by changing the steering angle, two steering angles are determined to bypass the obstacle.

[0129] Example 3

[0130] This is the third embodiment of the present invention. This embodiment provides a computer device, which is suitable for the case of an autonomous inspection and navigation method for a substation inspection robot based on laser radar, and includes a memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute computer executable instructions to implement the autonomous inspection and navigation method for a substation inspection robot based on laser radar as proposed in the above embodiment.

[0131] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0132] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the method for realizing an autonomous inspection and navigation method of a substation inspection robot based on a laser radar as proposed in the above embodiment is implemented.

[0133] The storage medium proposed in this embodiment and the data storage method proposed in the above embodiment belong to the same inventive concept. The technical details not fully described in this embodiment can be found in the above embodiment, and this embodiment has the same beneficial effects as the above embodiment.

[0134] Example 4

[0135] This is the second embodiment of the present invention. This embodiment provides an autonomous inspection and navigation method for a substation inspection robot based on laser radar. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation experiments.

[0136] In order to verify the effectiveness of the intelligent navigation system of the substation inspection robot based on laser radar and improved Dijkstra algorithm proposed in this invention, a field test was carried out in a 220kV substation. The substation covers an area of ​​about 10,000 square meters and contains a variety of high-voltage equipment, such as main transformers, circuit breakers, disconnectors, etc.

[0137] First, the Velodyne VLP-16 laser radar installed on the top of the inspection robot was used to collect environmental information. The laser radar has 16 laser channels, a 360-degree horizontal field of view, a 30-degree vertical field of view, and a measurement range of up to 100 meters. During the collection process, the robot was controlled to move along the preset path at a speed of 0.5m / s, while the laser radar scanned at a frequency of 10Hz. A total of about 15 minutes of data was collected, and raw point cloud data containing about 9 million points was obtained.

[0138] Next, the original point cloud data is processed using a graph-optimized SLAM algorithm to construct a high-precision 3D point cloud map. This algorithm is implemented using the g2o framework and uses a robust loop detection method to improve the accuracy of the map. Figure 1 The processed 3D point cloud map contains about 5 million points with centimeter-level accuracy.

[0139] Subsequently, the 3D point cloud map is converted into a 2D grid map. The height threshold is set to 1.8 meters (slightly higher than the robot height of 1.5 meters), and the points below the threshold are projected onto the 2D plane. The plane is divided into 0.1 meter × 0.1 meter grids, and the number of points in each grid is counted. If the number of points exceeds 50, the grid is marked as occupied, otherwise it is marked as free. Finally, a 3×3 median filter is used to smooth the grid map and remove noise. Finally, a 1000×1000 2D grid map with a resolution of 0.1 meters / pixel is obtained.

[0140] Based on the generated two-dimensional grid map, the improved Dijkstra algorithm was implemented. On the basis of the original algorithm, three factors, namely safety level, equipment importance and obstacle density, were introduced. The safety level function S(vi,vi+1) is defined according to the distance from the node to the nearest high-voltage equipment. The farther the distance, the higher the safety level. The equipment importance function E(vi,vi+1) is determined according to the predefined equipment importance level table. The obstacle density function O(vi,vi+1) is obtained by calculating the proportion of the grid occupied in the 3×3 area around the node.

[0141] In order to comprehensively evaluate the performance of this system, 6 groups of navigation tasks with different starting and ending points were designed, and the traditional Dijkstra algorithm and the improved algorithm of the present invention were used for path planning. At the same time, indicators such as planning time, path length, average safety distance (average distance to high-voltage equipment), number of important equipment passed, and obstacle avoidance effect (minimum distance to obstacles) were recorded. The test results are shown in the following table:

[0142] Table 1 Experimental data table

[0143]

[0144] By analyzing the data in the above table, we can draw the following conclusions:

[0145] Number of important equipment passed: The improved algorithm increased the coverage of important equipment by 66.7% on average. This means that under similar path lengths, the improved algorithm can arrange inspection routes more effectively and improve inspection efficiency.

[0146] Obstacle avoidance effect: The minimum distance between the path generated by the improved algorithm and the obstacle increased by 85% on average, from 0.72 meters to 1.33 meters. This not only improves the safety of navigation, but also increases the smoothness of the path, which is beneficial to the robot's motion control.

[0147] Path length: The paths generated by the improved algorithm are 7-10% longer than those of the traditional algorithm. This increase is expected because the improved algorithm not only considers distance, but also weighs safety, important equipment coverage, and obstacle avoidance. The slightly increased path length is exchanged for higher inspection quality and safety.

[0148] Average safety distance: The improved algorithm significantly improves the safety of the inspection path. The average safety distance has increased by about 60%, from 2.15 meters to 3.45 meters. This greatly reduces the risk of accidental contact between the inspection robot and high-voltage equipment.

[0149] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A laser radar-based substation inspection robot autonomous inspection navigation method, characterized in that: include, Collecting substation environmental information by laser radar to obtain a laser data point set, and constructing a two-dimensional grid map based on the laser data point set; After receiving the target location, the optimal path is planned using the two-dimensional grid map; the optimal path is defined based on the comprehensive distance, safety, equipment importance and obstacle density; According to the optimal path inspection, real-time environmental perception and self-positioning are performed to determine whether to enter the obstacle avoidance mode to achieve autonomous inspection.

2. The autonomous inspection and navigation method of a substation inspection robot based on a laser radar as claimed in claim 1, characterized in that: The constructing of the two-dimensional grid map is based on the three-dimensional point cloud map constructed by the laser data point set and converted into a two-dimensional grid map while retaining obstacle information; The conversion of 3D point cloud map to 2D raster map includes: A height threshold suitable for the substation environment is set, the entire three-dimensional point cloud data is traversed, and all points below the threshold are projected onto a two-dimensional plane. Then, the two-dimensional plane is divided into uniform grids, and each grid represents a unit in the grid map. Subsequently, the number of points in each grid is counted. If the number of points exceeds the preset threshold, the grid is marked as occupied, otherwise it is marked as free. Finally, the generated two-dimensional grid map is smoothed to remove isolated grids caused by noise.

3. The autonomous inspection and navigation method of a substation inspection robot based on laser radar as claimed in claim 2 is characterized by: The optimal path is planned based on comprehensive distance, safety, equipment importance, and obstacle density, as shown in the following formula: Among them, P opt is the optimal path, p is a possible path in the path set P, n is the number of nodes on the path, C( vi ,v i+1 ) is a comprehensive index combining various factors, v i and v i+1 is a node; The C( vi ,v i+1 ) is shown in the following formula: C(v i ,v i+1 )=C e (v i ,v i+1 )·(1+γ·O(v i ,v i+1 )) Among them, O( vi ,v i+1 ) is the obstacle density function between nodes, γ is the obstacle influence weight, C e ( vi ,v i+1 ) To ensure that important equipment indicators are met; The C e ( vi ,v i+1 ) is shown in the following formula: C e (in i ,v i+1 )=C s (in i ,v i+1 )·(1-β·E(v i ,v i+1 )) Among them, E( vi ,v i+1 ) is the device importance function between nodes, β is the importance weight, C s ( vi ,v i+1 ) To ensure safety indicators; The C s ( vi ,v i+1 ) is shown in the following formula: C s (in i ,in i+1 )=C d (in i ,in i+1 )·(1+a·S(v i ,in i+1 )) Among them, S( vi ,v i+1 ) is the security level function between nodes, α is the security level weight, C d ( vi ,v i+1 ) is the basic distance cost indicator to ensure the shortest path; The C d ( vi ,v i+1 ) is shown in the following formula: C d (v i ,v i+1 )=D(v i ,v i+1 ) Among them, D( vi ,v i+1 ) is the node v i and v i+1 The Euclidean distance between The O( vi ,v i+1 )、E( vi ,v i+1 )、S( vi ,v i+1 ) and D( vi ,v i+1 ) is shown in the following formula: S(v i ,v i+1 )=max(S i ,S i+1 ) / S max E(v i ,v i+1 )=(E i +E i+1 ) / (2·E max )' O(v i ,v i+1 )=count(obstacles) / max obstacles Among them, x i and i For node v i The coordinates of S i For node v i The security level, S max The highest security level, E i For node v i The importance value of nearby devices, E max is the highest importance value, count(obstacles) is the number of obstacles on the path segment, max obstacles The maximum number of obstacles that can be set.

4. The autonomous inspection and navigation method of a substation inspection robot based on laser radar as claimed in claim 3 is characterized by: The determination of whether to enter the obstacle avoidance mode includes: When the robot is inspecting in a substation environment, the system continuously uses lidar to scan the surrounding 360-degree environment, focusing on analyzing the data in front, on the left, and on the right, and determining the safety factor in front. At the same time, the system uses lidar data and point cloud processing algorithms to estimate the size of detected obstacles, compares the safety factor with a preset first threshold, and compares the estimated obstacle size with a preset second threshold. When the safety factor is lower than the first threshold or the obstacle size is greater than the second threshold, it further evaluates whether the obstacle interferes with the robot's normal travel path. If so, it enters an obstacle avoidance state, otherwise it enters a non-obstacle avoidance state.

5. The autonomous inspection and navigation method of a substation inspection robot based on laser radar as claimed in claim 4 is characterized in that: In the obstacle avoidance state, multiple candidate offset points are generated around the optimal path, and the best offset waypoint is selected by comprehensively evaluating the safety of each candidate point, the degree of deviation from the original path, and the impact on the subsequent path. After the offset waypoint is determined, a local path is quickly generated between the original global path and the new offset point; The determination of the offset waypoint is shown in the following formula: Among them, P offset is the best offset waypoint; C is the set of candidate offset points; α, β, γ are weight coefficients, and α+β+γ=1; F total (p) is the improved artificial potential field function; d(p,P original ) is the candidate point p and the original path point P original distance; I(p) is the influence factor of candidate point p on the subsequent path.

6. The autonomous inspection and navigation method of a substation inspection robot based on laser radar as claimed in claim 5, characterized in that: In the non-obstacle avoidance state, determine whether the obstacle is a dynamic obstacle. If so, predict the relative motion state between the obstacle in front and the robot based on the Kalman filter, and adjust the robot speed in real time through the adaptive cruise control algorithm in combination with the speed constraint conditions unique to the substation; If a static obstacle is detected, the system first evaluates the position, size and relationship of the obstacle to the robot's planned path. If the static obstacle is judged to interfere with the robot's normal path and cannot be circumvented by changing the steering angle, it switches to obstacle avoidance mode. If the static obstacle can be circumvented by changing the steering angle, two steering angles are determined to bypass the obstacle.

7. The autonomous inspection and navigation method of a substation inspection robot based on laser radar as claimed in claim 6, characterized in that: The two-stage steering angle is determined according to the obstacle area, channel width and distance, as shown in the following formula: Among them, θ base is the basic obstacle avoidance angle, w obs is the obstacle width, d safe is the safety distance, d obs is the distance from the obstacle to the robot, A obs is the obstacle area, A max is the preset maximum obstacle area, w channel is the channel width, R max is the maximum impact distance.

8. A laser radar-based substation inspection robot autonomous inspection navigation system, based on the laser radar-based substation inspection robot autonomous inspection navigation method according to any one of claims 1 to 7, characterized in that: include, A data acquisition module is used to collect substation environmental information through a laser radar to obtain a laser data point set, and to construct a two-dimensional grid map based on the laser data point set; A path planning module, used to plan an optimal path using the two-dimensional grid map after receiving the target location; the optimal path is defined by comprehensive distance, safety, equipment importance and obstacle density; The navigation control module is used to inspect according to the optimal path, perform real-time environmental perception and self-positioning, determine whether to enter the obstacle avoidance mode, and realize autonomous inspection.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the autonomous inspection and navigation method of a substation inspection robot based on laser radar as described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, the steps of the autonomous inspection and navigation method of the substation inspection robot based on laser radar as described in any one of claims 1 to 7 are implemented.

Images