A method, device and storage medium for robot indoor navigation

Abstract

The present invention proposes a method for indoor navigation of a robot, the method comprising: obtaining indoor environment information and obstacle information through sensors; classifying the obstacle information; formulating a coping strategy for each obstacle according to the classification results of the obstacles; the robot performs path planning on an established environment map according to a current position, a target position, the obstacle information and the coping strategy; the robot moves along the planned path, senses changes in the environment in real time, and dynamically adjusts the path according to new environmental information. Through this method, the robot can obtain and process surrounding environment information in a timely manner, the robot can perform more intelligent and optimized path planning, thereby providing more accurate navigation and obstacle avoidance, and improving navigation efficiency; by identifying obstacle types and selecting appropriate obstacle avoidance strategies, the robot can avoid collisions with obstacles, thereby improving navigation safety.

Description

Technical Field

[0001] The present invention relates to the field of indoor navigation technology, and in particular to a method, a device and a storage medium for indoor navigation of a robot. Background Art

[0002] Robotic indoor navigation is a key research area, which aims to enable robots to move autonomously in indoor environments and reach their target locations accurately. This technology is very important in many applications, such as smart homes, medical care, warehouse logistics, etc.

[0003] Robot indoor navigation methods need to provide higher navigation accuracy and flexibility. How to achieve efficient and flexible navigation and reduce collisions are problems that need to be solved. Summary of the invention

[0004] The present invention provides a method, device and storage medium for indoor navigation of a robot, which are used to provide more accurate navigation and obstacle avoidance and improve the efficiency and speed of navigation.

[0005] The present invention provides a method for indoor navigation of a robot, the method comprising:

[0006] S1, obtain indoor environment information and obstacle information through sensors;

[0007] S2. classifying the obstacle information; formulating a coping strategy for each obstacle according to the obstacle classification results;

[0008] S3, the robot performs path planning on the established environment map according to the current position, target position, obstacle information and response strategy;

[0009] S4. The robot moves along the planned path, perceives changes in the environment in real time, and dynamically adjusts the path based on the new environmental information.

[0010] Furthermore, a method for indoor navigation of a robot, wherein S1 comprises:

[0011] The robot obtains environmental information through sensors; the environmental information includes the location and shape data of indoor structures, furniture, and other obstacles; the sensors include laser radar, camera and infrared;

[0012] Based on the acquired environmental information, image processing and three-dimensional reconstruction technology are used to convert the perceived environmental data into an environmental map; the environmental map includes the location, shape, and size information of obstacles;

[0013] The map is divided into different grid areas; the area of ​​each grid area is:

[0014] =(1.5+ × ×S

[0015] Among them, S is the ground projection area of ​​the robot, and M is the number of obstacles per unit area; The number of obstacles preset per unit area; the actual navigation accuracy required, Preset navigation accuracy.

[0016] Furthermore, a method for indoor navigation of a robot, said S2 comprises:

[0017] Performing a first classification on the obstacles to obtain a first classification result, wherein the first classification result includes movable obstacles and immovable obstacles, and directly avoiding the immovable obstacles;

[0018] The movable obstacle is subjected to a second classification based on the first classification result to obtain a second classification result; the second classification is solid obstacles, liquid obstacles and moving obstacles; if it is a moving obstacle, the vehicle decelerates or stops to avoid it;

[0019] Based on the second classification result, the solid obstacles and the liquid obstacles are classified into a third category according to the size of the obstacles; the size of the obstacles is compared with the threshold value to obtain a comparison result;

[0020] If it is a solid obstacle, the length, width and height of the obstacle are compared with the first threshold; if it is an irregular obstacle, the length, width and height of the circumscribed rectangle are taken; first determine the height. If the height is greater than or equal to the robot's passable height, the robot will directly avoid it;

[0021] If the height is less than the robot's passable height, the length and width are compared; if the larger value of the length and width of the obstacle is less than or equal to the robot's passable threshold, the robot can pass directly; if the smaller value of the length and width of the obstacle is greater than the robot's passable threshold, the robot decelerates and / or avoids the obstacle; if the larger value of the length and width of the obstacle is greater than the robot's passable threshold, but the smaller value is less than the robot's passable threshold, the robot determines whether to avoid or pass the obstacle based on the robot's route and the position of the obstacle;

[0022] If it is a liquid obstacle, the area of ​​the obstacle is compared with the area threshold; if the area of ​​the liquid obstacle is smaller than the area threshold, the robot can pass directly; if the area of ​​the liquid obstacle is larger than the area threshold, the robot avoids it.

[0023] Furthermore, a method for indoor navigation of a robot, wherein if the larger value of the length and width of the obstacle is greater than the threshold value that the robot can pass through, but the smaller value is less than the threshold value that the robot can pass through, then avoiding or passing through is determined according to the robot's route and the position of the obstacle; comprising:

[0024] Determine whether to avoid the obstacle based on the angle θ between the obstacle and the robot's travel direction and its position;

[0025] If L× If it is greater than the threshold y that the robot can pass, avoid it;

[0026] If L× If it is less than or equal to the threshold for the robot to pass, it is considered passable;

[0027] If the distance b1 / b2 between the center point a of the robot's direction of travel and the end point projection of the obstacle is less than or equal to y / 2, the robot can pass directly; if any of the distances b1 / b2 between the center point a of the robot's direction of travel and the end point projection of the obstacle is greater than y / 2, the robot needs to adjust its posture to pass; the adjustment distance is D:

[0028] D=d1+V×(t1+t2)

[0029] D is the distance between the robot and the nearest point of the obstacle; d1 is the safety distance; V is the robot's travel speed; t1 is the adjustment time, and t2 is the sum of signal transmission and delay time.

[0030] Further, a method for indoor navigation of a robot, said S3 comprises:

[0031] The robot obtains current position information through sensors;

[0032] Generate multiple driving paths based on current location information, target location, grid information, and historical route information;

[0033] Calculate the distance in each path, the number of obstacles that need to be avoided, and the number of obstacles that are uncertain whether to avoid; select the optimal path based on the calculation results.

[0034] Furthermore, a method for indoor navigation of a robot, wherein the distance in each path and the number of obstacles to be avoided and the number of obstacles that are uncertain whether to be avoided are calculated; and the optimal path is selected according to the calculation results, comprising:

[0035] The number of stationary obstacles encountered in each path is calculated based on the obstacle classification results and historical records;

[0036] = +

[0037] ×

[0038]

[0039] in, The number of obstacles that must be avoided on a certain route from the robot's position to the target location in the historical records; the obstacles that must be avoided are immovable obstacles under the first category;

[0040] The number of uncertain obstacles that need to be avoided on a certain route from the robot's position to the target location in the historical records; the uncertain obstacles that need to be avoided are movable randomly appearing obstacles, including randomly appearing obstacles that need to be avoided and randomly appearing uncertain obstacles that need to be avoided;

[0041] (X = ) is the number of obstacles that need to be avoided. The probability of is the maximum probability value; λ is the average occurrence rate of the uncertain number of obstacles that need to be avoided, that is, the average number of occurrences of the uncertain number of obstacles that need to be avoided in the unit space, λ>0; The number of obstacles that need to be avoided is uncertain; is the number of uncertain obstacles that need to be avoided corresponding to the maximum probability value;

[0042] Calculate a comprehensive score C based on the quantity and route distance;

[0043] C = q × ( )

[0044] is the distance of a route; is the average distance of all routes; The average number of static obstacles encountered across all routes; ; is the weight parameter; q is the number of obstacles in motion;

[0045] The path with the smallest comprehensive score is determined as the optimal path.

[0046] Furthermore, a method for indoor navigation of a robot is characterized in that S4 comprises:

[0047] Obtain surrounding environment information in real time and upload it to the database system;

[0048] Extracting features from the surrounding environment information to identify the type of obstacle; selecting an obstacle avoidance strategy according to the type of obstacle;

[0049] Update the route model based on real-time monitoring information and optimize the route.

[0050] The present invention proposes an electrical device, comprising a memory, a processor, and a computer program stored in the memory and executable on the memory, wherein the processor executes the program to implement the method for indoor navigation of a robot.

[0051] The present invention provides a non-temporary computer-readable storage medium on which a computer program is stored. The program is executed by a processor to implement the method for indoor navigation of a robot.

[0052] Beneficial effects of the present invention: Through the method, device and storage medium for indoor navigation of a robot described in the present invention, the robot can timely obtain and process the surrounding environment information, thereby providing more accurate navigation and obstacle avoidance; based on the generated environmental map and fine grid division, the robot can perform more intelligent and optimized path planning. By analyzing the obstacle information of each grid area, the robot can select the best path, avoid obstacles, and seamlessly switch between different areas to improve navigation efficiency; by identifying the obstacle type and selecting the appropriate obstacle avoidance strategy, the robot can avoid collision with obstacles and improve navigation safety; through dynamic route optimization and path overlap optimization, the robot can select the best path and avoid possible obstacles, thereby improving navigation efficiency and speed; according to different environments and task requirements, the robot can flexibly adjust the navigation strategy and path planning to provide a personalized and adaptable navigation experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 This is a schematic diagram of a robot indoor navigation method according to the present invention;

[0054] Figure 2 Schematic diagram of the position relationship between the robot and the obstacle. DETAILED DESCRIPTION

[0055] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0056] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. The embodiments described are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0057] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used in the specification of the present invention herein are only for the purpose of describing specific embodiments and are not intended to limit the present invention.

[0058] This embodiment provides a method for indoor navigation of a robot, the method comprising:

[0059] S1, obtain indoor environment information and obstacle information through sensors;

[0060] S2. classifying the obstacle information; formulating a coping strategy for each obstacle according to the obstacle classification results;

[0061] S3, the robot performs path planning on the established environment map according to the current position, target position, obstacle information and response strategy;

[0062] S4. The robot moves along the planned path, perceives changes in the environment in real time, and dynamically adjusts the path based on the new environmental information.

[0063] The working principle of the above technical solution is as follows: the robot obtains indoor environment information and obstacle information through the sensors (camera, laser radar, infrared sensor, etc.) carried by it; the sensors can provide data such as the position, shape and distance of objects such as walls, furniture, and human bodies; the obstacle information obtained is classified and processed to distinguish different types of obstacles; for example, obstacles such as walls, furniture, and human bodies are classified. According to the classification results of the obstacles, corresponding response strategies are formulated. Each obstacle may require different avoidance or detour methods; according to the current position of the robot, the target position, the obstacle information and the response strategy, the path planning is carried out on the pre-established environmental map. The path planning algorithm can find an optimal or conditional path based on the robot's movement ability, the location of the obstacle and the limitations of the map; the robot moves along the planned path and perceives changes in the environment in real time. If a new obstacle appears or the position of a known obstacle changes, the robot will dynamically adjust the path according to the latest environmental information. This ensures that the robot maintains the correct navigation route in a real-time environment and avoids collisions with obstacles. This method obtains environmental information and obstacle information through sensors, classifies obstacles and formulates strategies, performs path planning, and perceives and dynamically adjusts the path in real time during movement, thereby realizing autonomous navigation of the robot in indoor environments.

[0064] The effect of the above technical solution is: by obtaining indoor environment information and obstacle information through sensors, high-precision perception of the environment can be achieved, thereby achieving a more accurate navigation effect. The robot can avoid collisions with obstacles and reach the target position safely. By classifying obstacles and formulating corresponding response strategies, different avoidance or detour methods can be adopted according to different types of obstacles. This enables the robot to make intelligent decisions when facing different obstacles and improves the flexibility and adaptability of navigation. By planning the path according to the current position, target position, obstacle information and response strategies, the robot can choose the optimal or conditional path. This can reduce the redundancy and time consumption of the path and improve the navigation efficiency; the robot perceives the changes in the environment in real time during the movement and dynamically adjusts the path according to the new environmental information. This can respond to new obstacles or changes in the position of known obstacles in a timely manner to ensure the accuracy and reliability of navigation.

[0065] In summary, the robot indoor navigation method can achieve precise navigation, adaptive strategy, path planning optimization and real-time dynamic adjustment, improving the robot navigation effect and user experience.

[0066] This embodiment provides a method for indoor navigation of a robot, wherein S1 includes:

[0067] The robot obtains environmental information through sensors; the environmental information includes the location and shape data of indoor structures, furniture, and other obstacles; the sensors include laser radar, camera and infrared;

[0068] Based on the acquired environmental information, image processing and three-dimensional reconstruction technology are used to convert the perceived environmental data into an environmental map; the environmental map includes the location, shape, and size information of obstacles;

[0069] The map is divided into different grid areas; the area of ​​each grid area is:

[0070] =(1.5+ × ×S

[0071] Among them, S is the ground projection area of ​​the robot, and M is the number of obstacles per unit area; The number of obstacles preset per unit area; the actual navigation accuracy required, Preset navigation accuracy.

[0072] The working principle of the above technical solution is as follows: the robot obtains indoor environmental information through sensors such as lidar, cameras and infrared. These sensors can perform distance measurement, image acquisition and object recognition on the environment to obtain detailed data of the surrounding environment. The robot processes the environmental information obtained by the sensor; this includes the position and shape data of indoor structures, furniture and other obstacles. Through image processing and three-dimensional reconstruction technology, the perceived environmental data is converted into an environmental map. In this way, the robot can have a more accurate and comprehensive understanding of the environment; in order to simplify the path planning and navigation process, the obtained environmental map will be divided into different grid areas. The area calculation formula for each grid area is:

[0073] =(1.5+ × ×S

[0074] Among them, S is the ground projection area of ​​the robot, M is the number of obstacles per unit area, is the preset number of obstacles per unit area, J is the actual required navigation accuracy, To preset navigation accuracy. Through grid division, complex environments can be transformed into discrete controllable areas.

[0075] In summary, the S1 part of this method obtains environmental information through sensors, performs image processing and 3D reconstruction, and generates an accurate environmental map. At the same time, the map is segmented by grid division to provide basic data for subsequent path planning and navigation. In this way, the robot can better perceive the environment and perform accurate indoor navigation.

[0076] The effect of the above technical solution is: the environmental information obtained by the sensor, including the location and shape data of indoor structures, furniture, and other obstacles, can generate an environmental map more accurately. This will provide accurate navigation data for the robot, avoiding blind movement and collision with obstacles; by dividing the map into different grid areas, the area of ​​each grid area is calculated according to the formula; this grid division method can adjust the size of each area according to the actual situation, making the navigation process more efficient and refined.

[0077] Based on the generated environment map and fine grid division, the robot can perform more intelligent and optimized path planning. By analyzing the obstacle information of each grid area, the robot can choose the best path, avoid obstacles, and switch seamlessly between different areas to improve navigation efficiency.

[0078] The formula takes into account two factors: the actual navigation accuracy requirement and the preset navigation accuracy. The higher the actual navigation accuracy requirement, the more detailed the grid division is required to meet the higher path planning accuracy. It serves as a benchmark reference to help determine the initial grid size. By considering the number of obstacles per unit area and the number of preset obstacles, the complexity of the environment can be taken into account when dividing the grid area; if the obstacle density in the environment is high, the obstacle distribution can be more accurately represented by increasing the number of grids. This formula can be adjusted according to the different robot ground projection areas S so that the size of the grid area matches the size of the robot; this helps to ensure that each grid can accommodate at least one projection of the robot to avoid misjudgments during path planning and obstacle avoidance; this formula can help comprehensively consider factors such as actual navigation accuracy, obstacle density and robot size when dividing the grid area, providing a basis for path planning and obstacle avoidance to achieve the preset navigation accuracy requirements; at the same time, the simplified path planning and obstacle avoidance algorithm reduces the computational complexity and consumption of computing resources compared to global planning and perception, and improves navigation efficiency.

[0079] In general, this robot indoor navigation method obtains environmental information through sensors and performs image processing and three-dimensional reconstruction to generate an accurate environmental map, and adopts grid division and optimized path planning to provide a more efficient and accurate indoor navigation experience.

[0080] This embodiment provides a method for indoor navigation of a robot, wherein S2 includes:

[0081] Obstacles are first classified to obtain a first classification result, wherein the first classification result includes movable obstacles and immovable obstacles; the immovable obstacles include walls, solid pillars, fences, cabinets, and door stops; movable obstacles include tables, chairs, mobile carts, people, animals, sundries, water stains, etc.; the immovable obstacles are directly avoided;

[0082] The movable obstacles are subjected to a second classification based on the first classification result to obtain a second classification result; the second classification includes solid obstacles, liquid obstacles and moving obstacles; the moving obstacles include people, animals and other moving objects;

[0083] Based on the second classification result, the solid obstacles and the liquid obstacles are classified into a third category according to the size of the obstacles; the size of the obstacles is compared with the threshold value to obtain a comparison result;

[0084] If it is a solid obstacle, the length, width and height of the obstacle are compared with the first threshold; if it is an irregular obstacle, the length, width and height of the circumscribed rectangle are taken; first determine the height. If the height is greater than or equal to the robot's passable height, the robot will directly avoid it;

[0085] If the height is less than the robot's passable height, the length and width are compared; if the larger value of the length and width of the obstacle is less than or equal to the robot's passable threshold, the robot can pass directly; if the smaller value of the length and width of the obstacle is greater than the robot's passable threshold, the robot decelerates and / or avoids the obstacle; if the larger value of the length and width of the obstacle is greater than the robot's passable threshold, but the smaller value is less than the robot's passable threshold, the robot determines whether to avoid or pass the obstacle based on the robot's route and the position of the obstacle;

[0086] If it is a liquid obstacle, the area of ​​the obstacle is compared with the area threshold; if the area of ​​the liquid obstacle is smaller than the area threshold, the robot can pass directly; if the area of ​​the liquid obstacle is larger than the area threshold, the robot avoids it;

[0087] If it is a moving obstacle, slow down or stop to avoid it.

[0088] The working principle of the above technical solution is as follows: First classification: obstacles are classified into two categories: movable obstacles and immovable obstacles. Immovable obstacles include walls, solid pillars, fences, cabinets, doors, etc., which need to be avoided directly. Movable obstacles include tables, chairs, mobile carts, people, animals, debris, water stains, etc.

[0089] Second classification: further classify movable obstacles into solid obstacles, liquid obstacles and moving obstacles. Moving obstacles include people, animals and other moving objects. For moving obstacles: the robot will slow down or stop to avoid moving obstacles to ensure safe passage.

[0090] Third classification: solid obstacles and liquid obstacles are further classified according to the size of the obstacles. The type of obstacle is determined by comparing the size of the obstacle with the threshold.

[0091] For solid obstacles: compare the length, width, and height of the obstacle with the first threshold. If the height of the obstacle is greater than or equal to the robot's passable height, the robot will directly avoid it. If the height of the obstacle is less than the robot's passable height, compare the length and width of the obstacle. If the larger value of the obstacle is less than or equal to the robot's passable threshold, it can pass directly. If the smaller value of the obstacle is greater than the robot's passable threshold, slow down and / or avoid it. If the larger value of the obstacle is greater than the robot's passable threshold, but the smaller value is less than the robot's passable threshold, determine the avoidance or passage method based on the robot's route and the position of the obstacle.

[0092] For liquid obstacles: compare the area of ​​the obstacle with the area threshold. If the area of ​​the liquid obstacle is smaller than the threshold, the robot can pass through it directly. If the area of ​​the liquid obstacle is larger than the threshold, the robot avoids it.

[0093] In summary, this method enables the robot to judge the type and size of obstacles through multi-level classification and threshold comparison, and take corresponding avoidance or passage strategies as needed, thereby achieving the safety and effectiveness of indoor navigation.

[0094] The effect of the above technical solution is: by classifying and judging obstacles, the robot can accurately identify immovable obstacles and avoid them directly, avoid collisions with fixed obstacles such as walls and pillars, and ensure safety during navigation. The method further classifies movable obstacles, distinguishes them according to the size of the obstacles, and combines threshold comparison to enable the robot to quickly decide whether to avoid or pass directly. This can reduce the number of pauses and adjustments of the robot and improve the efficiency of navigation. Taking into account the characteristics of different types of obstacles, when dealing with solid obstacles, not only the height but also the length and width are considered, so that the robot can adapt to solid obstacles of different sizes. At the same time, liquid obstacles and moving obstacles are also handled accordingly, which improves the robot's adaptability in complex environments. According to the location of the obstacle and the robot's route, the method can flexibly select avoidance or passage strategies to ensure that the robot can safely navigate in narrow or complex spaces, thereby improving the flexibility and adaptability of the robot.

[0095] In summary, the robot indoor navigation method improves the safety, efficiency, adaptability and flexibility of navigation through classification and threshold comparison, and can better cope with various indoor navigation scenarios.

[0096] This embodiment provides a method for indoor navigation of a robot, wherein if the larger value of the length and width of an obstacle is greater than a threshold value that the robot can pass through, but the smaller value is less than the threshold value that the robot can pass through, then avoiding or passing through is determined according to the robot's route and the position of the obstacle; the method comprises:

[0097] Determine whether to avoid the obstacle based on the angle θ between the obstacle and the robot's direction of travel and its position; Figure 2 As shown:

[0098] If L× If it is greater than the threshold y that the robot can pass, avoid it;

[0099] If L× If it is less than or equal to the threshold for the robot to pass, it is considered passable;

[0100] If the distance b1 / b2 between the center point a of the robot's direction of travel and the end point projection of the obstacle is less than or equal to y / 2, the robot can pass directly; if any of the distances b1 / b2 between the center point a of the robot's direction of travel and the end point projection of the obstacle is greater than y / 2, the robot needs to adjust its posture to pass; the adjustment distance is D:

[0101] D=d1+V×(t1+t2)

[0102] D is the distance between the robot and the nearest point of the obstacle; d1 is the safety distance; V is the robot's travel speed; t1 is the adjustment time, and t2 is the sum of signal transmission and delay time.

[0103] The working principle of the above technical solution is: determine the angle θ and position between the obstacle and the robot's travel direction: obtain the position information of the obstacles around the robot through sensors or visual systems, and calculate the angle θ between the obstacle and the robot's travel direction. According to the comparison between the formula L×cos(θ) (L is the larger value of the obstacle length and width) and the threshold y that the robot can pass, determine whether it is necessary to avoid. If L×cos(θ) is greater than the threshold y that the robot can pass, it means that the obstacle is on the robot's travel path and needs to be avoided; if L×cos(θ) is less than or equal to the threshold y that the robot can pass, it means that the size of the obstacle will not hinder the robot's passage and can pass directly; if any distance between the center point a of the robot's travel direction and the projection distance b1 / b2 of the end point of the obstacle is greater than y / 2, it is necessary to adjust the posture to pass to avoid collision with the obstacle. The adjusted distance D is calculated according to the formula, where d1 is the safety distance, V is the robot's travel speed, t1 is the adjustment time, and t2 is the sum of the signal transmission and delay time.

[0104] The effect of the above technical solution is: by judging the size of the obstacle and the robot's passage threshold, the robot can effectively avoid collision or obstruction with large obstacles. When the size of the obstacle is large, the robot can avoid it in time to ensure safe passage; in the case of direct passage, the robot does not need to perform additional avoidance actions, which can reduce the time of path planning and motion adjustment and improve navigation efficiency. Only when avoidance is required, the posture is appropriately adjusted to minimize the robot's route and time; this method takes into account the angle between the obstacle and the robot's travel direction, and determines whether avoidance is required according to the specific situation. Therefore, when dealing with obstacles of different types and positions, decisions can be made flexibly and the optimal navigation strategy can be selected. This method calculates the robot's route and adjustment distance in real time based on the current position and size of the obstacle. Through instant response and dynamic planning, it can cope with environmental changes and obstacle movement to ensure the accuracy and real-time performance of navigation.

[0105] Based on the relationship between the size of the obstacle, the angle between the robot's direction of travel and the position, a simple and clear judgment and decision-making method is provided. By directly comparing the product of the length and angle with the size of the threshold that the robot can pass, it can be quickly determined whether to avoid or pass directly. The parameters and variables in the formula can be adjusted according to actual needs to adapt to the robot navigation needs in different scenarios. For example, the adjustment distance can be calculated based on the robot's travel speed and safety distance, so that the robot can remain safe and stable when adjusting its posture. By judging the angle and position between the obstacle and the robot's travel direction, it is possible to flexibly decide in a real-time environment whether to avoid or pass directly. This can quickly respond to environmental changes and ensure the real-time and flexibility of the navigation process. The adjustment distance in the formula takes into account factors such as the robot's travel speed and delay time. By calculating the adjustment distance according to actual conditions, the adjustment of the robot's travel path can be minimized to improve navigation efficiency.

[0106] In summary, the robot indoor navigation method judges the obstacle size and the robot's passable threshold, flexibly selects avoidance or direct passage strategies, and adjusts the posture when necessary to improve safety, efficiency and flexibility, thereby achieving high-quality indoor navigation.

[0107] This embodiment provides a method for indoor navigation of a robot, wherein S3 includes:

[0108] The robot obtains current position information through sensors;

[0109] Generate multiple driving paths based on current location information, target location, grid information, and historical route information;

[0110] Calculate the distance in each path, the number of obstacles that need to be avoided, and the number of obstacles that are uncertain whether to avoid; select the optimal path based on the calculation results.

[0111] The working principle of the above technical solution is as follows: the robot obtains the current position information through the sensor, such as using a laser radar, a camera or other positioning sensor to determine the position of the robot; the target position of the robot is determined according to the task requirements or user settings. At the same time, the environment is divided into grids for path planning and calculation of obstacle avoidance; based on the current position information, the target position and the grid information, a path planning algorithm (such as the A* algorithm, the Dijkstra algorithm, etc.) is used to generate multiple driving paths. The path planning algorithm considers the robot's passable area and the location of known obstacles to find the shortest path connecting the starting point and the end point; for each generated path, its length is calculated as an evaluation index of the distance. At the same time, by analyzing the obstacle information on the path, the number of obstacles that need to be avoided and the number of obstacles that are uncertain whether to avoid are counted. For example, it can be determined whether to avoid by judging the size of the obstacle and the threshold of the robot's passability; according to the distance of the path and the number of obstacles to avoid, combined with historical route information (such as the number of times passed, time, etc.), the optimal path is selected through an evaluation function or optimization algorithm. The evaluation function can be designed according to specific needs, such as weighing the weight of the distance and the number of obstacles to avoid.

[0112] Through the above steps, the robot can obtain the current location information and generate multiple driving paths based on the target location, grid information and historical route information. After calculating the distance of each path and the number of obstacles to be avoided, the optimal path is selected to achieve indoor navigation of the robot. This method can provide a variety of path options and take into account the factors of distance and obstacle avoidance to improve the effect and safety of navigation.

[0113] The effect of the above technical solution is: by generating multiple driving paths according to the current position information, the target position, the grid information and the historical route information, the robot can have multiple choices during the navigation process, thereby improving the flexibility and adaptability of the path. In this way, it can cope with different environmental changes and obstacle distribution conditions and improve the success rate of navigation. By calculating the distance in each path, the robot can choose the shortest path to reach the target position. This can reduce the robot's travel time and energy consumption and improve the efficiency of navigation. Calculating the number of obstacles that need to be avoided in the path and the number of obstacles that are uncertain whether to be avoided can help the robot avoid collisions and safety issues during navigation. By avoiding obstacles, the safety of the robot and the environment can be protected and the risk of accidents can be reduced. Combined with historical route information, considering the number of times the route passes, time and other factors when selecting the optimal path can help the robot choose a more reliable and high-quality path. In this way, the success rate of navigation and user satisfaction can be improved based on past experience and feedback. By using sensors to obtain current position information and planning the path based on real-time grid information, the accuracy of navigation can be improved. The robot can better perceive the surrounding environment, adjust the path in time and avoid obstacles, so as to reach the target location more accurately.

[0114] Overall, this robot indoor navigation method can improve the flexibility, efficiency, safety and accuracy of navigation by means of multi-path selection, distance optimization, obstacle avoidance and consideration of historical route information. This will help robots complete navigation tasks efficiently and safely in indoor environments.

[0115] This embodiment provides a method for indoor navigation of a robot, wherein the distance in each path and the number of obstacles to be avoided and obstacles for which it is uncertain whether to avoid are calculated; and the optimal path is selected according to the calculation results, including:

[0116] The number of stationary obstacles encountered in each path is calculated based on the obstacle classification results and historical records;

[0117] = +

[0118] ×

[0119]

[0120] in, The number of obstacles that must be avoided on a certain route from the robot's position to the target location in the historical records; the obstacles that must be avoided are immovable obstacles under the first category;

[0121] The number of uncertain obstacles that need to be avoided on a certain route from the robot's position to the target location in the historical records; the uncertain obstacles that need to be avoided are movable randomly appearing obstacles, including randomly appearing obstacles that need to be avoided and randomly appearing uncertain obstacles that need to be avoided;

[0122] (X = ) is the number of obstacles that need to be avoided. The probability of is the maximum probability value; λ is the average occurrence rate of the uncertain number of obstacles that need to be avoided, that is, the average number of occurrences of the uncertain number of obstacles that need to be avoided in the unit space, λ>0; The number of obstacles that need to be avoided is uncertain; is the number of uncertain obstacles that need to be avoided corresponding to the maximum probability value;

[0123] Calculate a comprehensive score C based on the quantity and route distance;

[0124] C = q × ( )

[0125] is the distance of a route; is the average distance of all routes; The average number of static obstacles encountered across all routes; , is the weight parameter, and its value range is (0, 1); q is the number of obstacles in motion;

[0126] The path with the smallest comprehensive score is determined as the optimal path.

[0127] The working principle of the above technical solution is: obstacle classification and quantity calculation: according to the classification results and historical records of obstacles, the number of stationary obstacles encountered in each path is calculated. Indicates the number of obstacles that must be avoided on a route from the robot's position to the target location in the historical records. These obstacles belong to the immovable obstacles in the first category; It indicates the number of uncertain obstacles that need to be avoided on a route from the robot's position to the target location in the historical records. These obstacles are movable and randomly appearing obstacles.

[0128] Use the probability distribution function to calculate the probability that the number of obstacles to be avoided is k (X= ), where λ represents the average occurrence rate of uncertain obstacles to be avoided, ki represents the number of uncertain obstacles to be avoided; max( ) represents the maximum probability value.

[0129] Calculate the comprehensive score C based on the number and distance of paths;

[0130] Optimal path selection: According to the size of the comprehensive score C, select the path with the smallest comprehensive score as the optimal path.

[0131] The effect of the above technical solution is: the method comprehensively evaluates the pros and cons of the path by considering multiple factors such as path distance, the number of static obstacles encountered, and the number of randomly occurring obstacles that need to be avoided. Therefore, it can more comprehensively consider the safety and efficiency of navigation; the method uses the obstacle classification results and numbers in the historical records to calculate the number of static obstacles encountered in the path. This means that the robot can avoid known obstacles based on past experience, thereby improving the accuracy and reliability of navigation; by using the probability distribution function, the method can calculate the probability of the number of uncertain obstacles to be avoided, and select the case with the highest probability as a reference. Doing so can better cope with randomly occurring movable obstacles and improve the adaptability and flexibility of navigation. By comprehensively considering factors such as path distance and the number of obstacles encountered, the method calculates a comprehensive score and selects the path with the smallest comprehensive score as the optimal path. In this way, various factors can be comprehensively balanced during navigation to find the safest and most efficient navigation path.

[0132] By using a formulaic approach, the navigation problem can be transformed into a mathematical model to accurately describe the relationship between factors such as path distance and the number of obstacles. This can avoid the influence of subjective judgment and uncertainty on the navigation results and improve the accuracy of navigation. The formulaic representation allows each factor to be quantified and their corresponding weights to be assigned. In this way, when considering multiple factors comprehensively, they can be objectively evaluated and compared to find the optimal navigation path. Through the formulaic representation, the parameters and weights in the formula can be adjusted according to actual needs to adapt to different navigation scenarios and user needs; if more attention needs to be paid to the short-range nature of the path, w1 can be increased; if more attention needs to be paid to avoiding obstacles, w2 can be increased. This makes the navigation method more flexible and can adapt to changes and special requirements in different environments; q in the formula represents the number of obstacles encountered during movement, which will further affect the comprehensive score; this means that when selecting the optimal path, not only the number of static obstacles on the path should be considered, but also the number of additional obstacles that may appear during the movement, thereby improving the safety and robustness of the path; the comprehensive score C represents the comprehensive evaluation of the path. By comparing the comprehensive scores of different paths, the path with the smallest comprehensive score can be selected as the optimal path. In this way, the best path can be selected from multiple alternative paths to improve mobility efficiency and safety. The formulated navigation method can obtain the result through mathematical calculation, avoiding the trial and error and repeated attempts in the traditional method. This can save time and resources and improve navigation efficiency.

[0133] In summary, this robot indoor navigation method provides a more comprehensive, accurate and reliable navigation solution by considering obstacles and path information in multiple dimensions. It can improve the safety and efficiency of navigation, adapt to the randomness in different scenarios, and make intelligent decisions based on historical records, providing users with a better navigation experience.

[0134] This embodiment provides a method for indoor navigation of a robot, wherein S4 includes:

[0135] Obtain surrounding environment information in real time and upload it to the database system;

[0136] Extracting features from the surrounding environment information to identify the type of obstacle; selecting an obstacle avoidance strategy according to the type of obstacle;

[0137] Update route models based on real-time monitoring information and optimize routes;

[0138] According to the overlap between the actual path and the planned path, the weight value is optimized and adjusted.

[0139] The working principle of the above technical solution is as follows: the robot obtains information about the surrounding environment in real time through sensors (such as lidar, camera, etc.), including ground flatness, wall position, obstacle position, etc. This information is uploaded to the database system for subsequent processing and analysis. The uploaded environmental information is extracted from the database system, and the obstacles are feature extracted and classified. This can be achieved through image processing, deep learning and other technologies. By identifying different types of obstacles, such as furniture, people, etc., adaptive obstacle avoidance strategies can be better selected; according to the identified obstacle types, the robot can pre-define a series of obstacle avoidance strategies. Based on the information in the strategy library, the robot can decide how to bypass or avoid different types of obstacles to ensure safe and efficient navigation. The robot uses sensors to monitor changes in its own position and surrounding environment in real time. Based on this information, the robot can continuously update the current route model and optimize the path. This can be achieved through path planning algorithms, real-time positioning technology, etc., to ensure that the robot is always on the best path during navigation. By comparing the overlap between the actual path and the planned path, the current navigation effect can be evaluated. If it is found that there is a large deviation between the planned path and the actual path, the weight value in the path planning algorithm can be gradually optimized and adjusted according to the deviation. This can improve the accuracy and adaptability of path planning to better meet navigation needs.

[0140] The effect of the above technical solution is that the robot can obtain and process the surrounding environment information in time, so as to provide more accurate navigation and obstacle avoidance. By identifying the obstacle type and selecting the appropriate obstacle avoidance strategy, the robot can avoid collision with obstacles and improve the safety of navigation; through dynamic route optimization and path overlap optimization, the robot can select the best path and avoid possible obstacles, so as to improve the efficiency and speed of navigation; according to different environments and task requirements, the robot can flexibly adjust the navigation strategy and path planning to provide a personalized and adaptable navigation experience.

[0141] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A method for indoor navigation of a robot, characterized in that: The method comprises: S1, obtain indoor environment information and obstacle information through sensors; S2. classifying the obstacle information; formulating a coping strategy for each obstacle according to the obstacle classification results; S3, the robot performs path planning on the established environment map according to the current position, target position, obstacle information and response strategy; Wherein, the S3 includes: The robot obtains current position information through sensors; Generate multiple driving paths based on current location information, target location, grid information, and historical route information; Calculate the distance in each path, the number of obstacles that need to be avoided, and the number of obstacles that are uncertain whether to avoid. Select the optimal path based on the calculation results, including: The number of stationary obstacles encountered in each path is calculated based on the obstacle classification results and historical records; = + × in, The number of obstacles that must be avoided on a certain route from the robot's position to the target location in the historical records; the obstacles that must be avoided are immovable obstacles under the first category; The number of uncertain obstacles that need to be avoided on a certain route from the robot's position to the target location in the historical records; the uncertain obstacles that need to be avoided are movable randomly appearing obstacles, including randomly appearing obstacles that need to be avoided and randomly appearing uncertain obstacles that need to be avoided; (X = ) is the number of obstacles that need to be avoided. The probability of is the maximum probability value; λ is the average occurrence rate of the uncertain number of obstacles that need to be avoided, that is, the average number of occurrences of the uncertain number of obstacles that need to be avoided in the unit space, λ > 0; The number of obstacles that need to be avoided is uncertain; is the number of uncertain obstacles that need to be avoided corresponding to the maximum probability value; Calculate a comprehensive score C based on the quantity and route distance; C=q×( ) is the distance of a route; is the average distance of all routes; The average number of static obstacles encountered across all routes; ; is the weight parameter; q is the number of obstacles in motion; The path with the smallest comprehensive score is determined as the optimal path; S4. The robot moves along the planned path, perceives changes in the environment in real time, and dynamically adjusts the path based on the new environmental information.

2. A robot indoor navigation method according to claim 1, characterized in that: The S1 includes: The robot obtains environmental information through sensors; the environmental information includes the location and shape data of indoor structures, furniture, and other obstacles; the sensors include laser radar, camera and infrared; Based on the acquired environmental information, image processing and three-dimensional reconstruction technology are used to convert the perceived environmental data into an environmental map; the environmental map includes the location, shape, and size information of obstacles; The map is divided into different grid areas; the area of ​​each grid area is: =（1.5+ × ×S Among them, S is the ground projection area of ​​the robot, and M is the number of obstacles per unit area; The number of obstacles preset per unit area; the actual navigation accuracy required, Preset navigation accuracy.

3. A robot indoor navigation method according to claim 1, characterized in that: The S2 includes: Performing a first classification on the obstacles to obtain a first classification result, wherein the first classification result includes movable obstacles and immovable obstacles, and directly avoiding the immovable obstacles; The movable obstacle is subjected to a second classification based on the first classification result to obtain a second classification result; the second classification is solid obstacles, liquid obstacles and moving obstacles; if it is a moving obstacle, the vehicle decelerates or stops to avoid it; Based on the second classification result, the solid obstacles and the liquid obstacles are classified into a third category according to the size of the obstacles; the size of the obstacles is compared with the threshold value to obtain a comparison result; If it is a solid obstacle, the length, width and height of the obstacle are compared with the first threshold; if it is an irregular obstacle, the length, width and height of the circumscribed rectangle are taken; first determine the height. If the height is greater than or equal to the robot's passable height, the robot will directly avoid it; If the height is less than the robot's passable height, the length and width are compared; if the larger value of the length and width of the obstacle is less than or equal to the robot's passable threshold, the robot can pass directly; if the smaller value of the length and width of the obstacle is greater than the robot's passable threshold, the robot decelerates and / or avoids the obstacle; if the larger value of the length and width of the obstacle is greater than the robot's passable threshold, but the smaller value is less than the robot's passable threshold, the robot determines whether to avoid or pass the obstacle based on the robot's route and the position of the obstacle; If it is a liquid obstacle, the area of ​​the obstacle is compared with the area threshold; if the area of ​​the liquid obstacle is smaller than the area threshold, the robot can pass directly; if the area of ​​the liquid obstacle is larger than the area threshold, the robot avoids it.

4. A robot indoor navigation method according to claim 3, characterized in that: If the larger value of the length and width of the obstacle is greater than the threshold value that the robot can pass through, but the smaller value is less than the threshold value that the robot can pass through, then determine whether to avoid or pass through according to the robot's route and the position of the obstacle; including: Determine whether to avoid the obstacle based on the angle θ between the obstacle and the robot's travel direction and its position; If L× If it is greater than the threshold y that the robot can pass, avoid it; If L× If it is less than or equal to the threshold for the robot to pass, it is considered passable; If the distance b1 / b2 between the center point a of the robot's direction of travel and the end point projection of the obstacle is less than or equal to y / 2, the robot can pass directly; if any of the distances b1 / b2 between the center point a of the robot's direction of travel and the end point projection of the obstacle is greater than y / 2, the robot needs to adjust its posture to pass; the adjusted distance is D: D=d1+V×(t1+t2) D is the distance between the robot and the nearest point of the obstacle; d1 is the safety distance; V is the robot's travel speed; t1 is the adjustment time, and t2 is the sum of signal transmission and delay time.

5. The method for indoor navigation of a robot according to claim 1, characterized in that: The S4 includes: Obtain surrounding environment information in real time and upload it to the database system; Extracting features from the surrounding environment information to identify the type of obstacle; selecting an obstacle avoidance strategy according to the type of obstacle; Update the route model based on real-time monitoring information and optimize the route.

6. An electrical device, characterized in that: It comprises a memory, a processor and a computer program stored in the memory and executable on the memory, wherein the processor executes the program to implement a method for indoor navigation of a robot as described in any one of claims 1 to 5.

7. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: The program is executed by a processor to implement a robot indoor navigation method as described in any one of claims 1-5.

Images