Automatic campus patrol robot based on visual SLAM

Abstract

The utility model discloses an automatic campus patrol robot based on visual SLAM (Simultaneous Localization and Mapping), relates to the technical field of automatic robots, and solves the problems of high manufacturing cost, poor convenience, excessive functions and limited adaptability of the automatic robots. Collected image data are transmitted to the Raspberry Pi, map construction is carried out on the collected image data and information collected by the laser radar, and an infrared temperature measurement dot matrix sensor and a gas sensor are used for collecting temperature and humidity information and air and gas component information. The system automatically starts a voice alarm and reports data to workers, and the campus patrol robot is simple in structure, small in size, light in weight, suitable for a campus environment where people are dense, capable of running on complex terrains by adopting crawler walking, good in stability, capable of replacing human beings to work in a dangerous or harmful environment and capable of working for a long time.

Description

Technical Field

[0001] The utility model relates to the technical field of automated robots, in particular to an automated campus patrol robot based on visual SLAM. Background Art

[0002] As one of the representatives of modern technology, automated robots have been widely used in industrial manufacturing, service industry, medical treatment, agriculture and other fields around the world.

[0003] At present, the automation robot technology has made significant progress in sensor technology, navigation and positioning, artificial intelligence and machine learning, control systems, etc. These technological advances provide strong support for the application of robots in various environments, enabling automation robots to complete multiple tasks. However, most of the existing products are expensive and bulky. Most of them are suitable for use in the engineering field and require relevant professionals to complete related tasks. They lack daily practicality and cannot be popularized and promoted in daily life. The goal of popularizing and using them for daily management and patrol on campus has not yet been achieved.

[0004] Existing automated robot technology has achieved remarkable results in terms of functional variety, innovation and safety, meeting users' requirements for robot functions and safety. However, it has obvious shortcomings in terms of practicality and convenience in daily campus life. Its advantages and disadvantages are mainly reflected in the following aspects:

[0005] advantage:

[0006] Improve production efficiency: Automated robots can work for a long time, which significantly improves production efficiency and operation speed. In addition, automated robots can accurately and quickly detect problems and respond in time, reducing human errors.

[0007] Improve safety: Robots can replace humans in dangerous or harmful environments and monitor environmental anomalies, thereby reducing safety risks and hidden dangers in the workplace.

[0008] High flexibility: Modern automated robots are usually highly flexible and programmable, and can adapt to different production needs and working environments. Through programming or artificial intelligence, robots can learn new tasks and skills to meet changing market needs.

[0009] shortcoming:

[0010] Too high a cost: Existing automated robots have comprehensive functions but most of them are expensive and most companies cannot afford them.

[0011] Low convenience: Most automated robotic devices are large in size and are suitable for use in engineering fields and require relevant professionals to complete related tasks. For densely populated environments such as campuses, traditional automated robots are not convenient to operate.

[0012] Over-function: Automated robots have a variety of functions, but in actual applications, most of them only require a few of them, and the other functions of the robots are not well used, which results in over-function of the robots.

[0013] Adaptability limitations: Despite significant advances in modern robotics, robots still have limited adaptability and flexibility in handling complex, non-standardized, or non-routine tasks and are easily disrupted by environmental changes. Utility Model Content

[0014] In view of the problems of high cost, poor convenience, redundant functions and limited adaptability of the above-mentioned automated robots, the purpose of the utility model is to provide an automated campus patrol robot based on visual SLAM.

[0015] In order to achieve the above purpose, the technical solution adopted by the utility model is:

[0016] An automated campus patrol robot based on visual SLAM, comprising:

[0017] A crawler traveling mechanism 2 and a traveling chassis 1, wherein two crawler traveling mechanisms 2 are symmetrically installed on the left and right sides of the traveling chassis 1;

[0018] A layered support 4, wherein the layered support 4 is installed on the top of the walking chassis 1;

[0019] A gas sensor 9 and an infrared temperature measurement dot matrix sensor 8, wherein the gas sensor 9 and the infrared temperature measurement dot matrix sensor 8 are both mounted on a layered bracket 4;

[0020] A laser radar 3, wherein the laser radar 3 is installed on the top of a layered support 4;

[0021] A single servo and a depth camera 7, wherein the single servo is mounted on a layered bracket 4, and the depth camera 7 is mounted on the single servo;

[0022] The data processing device 5 is installed in the layered bracket 4, and the laser radar 3, the gas sensor 9, the infrared temperature measurement dot matrix sensor 8, the driving servo and the depth camera 7 are all connected to the data processing device by signal.

[0023] The above-mentioned automated campus patrol robot based on visual SLAM, wherein the layered bracket 4 includes: a top plate 43, a middle plate 42, a bottom plate 41, a first leg 45 and a second leg 46, the lower surface of the bottom plate 41 is connected to the top of the walking chassis 1, each corner of the middle plate 42 is equipped with a second leg 46, and the lower ends of multiple second legs 46 are connected to the upper surface of the bottom plate 41, and each corner of the lower surface of the top plate 43 is equipped with a first leg 45, and the lower ends of multiple first legs 45 are connected to the upper surface of the middle plate 42.

[0024] The above-mentioned automated campus patrol robot based on visual SLAM, wherein the layered bracket 4 also includes: a radar mounting bracket 44, the radar mounting bracket 44 is installed on the upper surface of the top plate 43, and the laser radar 3 is installed in the radar mounting bracket 44.

[0025] In the above-mentioned automated campus patrol robot based on visual SLAM, the data processing device 5 is installed on the middle layer board 42 .

[0026] The above-mentioned automated campus patrol robot based on visual SLAM, wherein the track walking mechanism 2 includes: a track bracket 23, a driving wheel 24, a driven wheel 25, a tensioning mechanism 26, a tensioning wheel 27 and a track 22, the track bracket 23 is installed on the walking chassis 1, and a plurality of driven wheels 25 are rotatably installed at the bottom of the track bracket 23, the tensioning mechanism 26 is installed on the front side of the track bracket 23, the driving wheel 24 is installed on the rear side of the track bracket, the tensioning wheel 27 is rotatably installed on the tensioning mechanism 26, and the track 22 is arranged around the driving wheel 24, the tensioning wheel 27 and the plurality of driven wheels 25.

[0027] In the above-mentioned automated campus patrol robot based on visual SLAM, the crawler walking mechanism 2 further includes: a DC motor 21 , the DC motor 21 is mounted on a crawler bracket 23 , and the driving wheel 24 is mounted at the output end of the DC motor 21 .

[0028] The above-mentioned automated campus patrol robot based on visual SLAM, wherein the crawler walking mechanism 2 also includes: a motor driver and a controller, the motor driver is installed on the walking chassis 1, the controller is installed on the layered bracket 4, the motor driver and the controller are connected through lines, and the motor driver and the DC motor 21 are connected through lines.

[0029] The above-mentioned automated campus patrol robot based on visual SLAM, among other things, also includes: a power supply device, the crawler walking mechanism 2, the laser radar 3, the data processing device 5, the depth camera 7, the infrared temperature measurement dot matrix sensor 8 and the gas sensor 9 are all connected to the power supply device through a circuit.

[0030] The above-mentioned automated campus patrol robot based on visual SLAM, wherein the data processing device includes: a single-chip microcomputer and a Raspberry Pi, the single-chip microcomputer and the Raspberry Pi are both connected to the power supply device through a circuit, and the single-chip microcomputer and the Raspberry Pi are signal-connected.

[0031] The above-mentioned automated campus patrol robot based on visual SLAM, among other things, also includes: a remote control device, the remote control device and the data processing device are connected by wireless signals, and the remote control device is provided with a display screen and an alarm device.

[0032] Due to the adoption of the above technology, the utility model has the following positive effects compared with the prior art:

[0033] (1) In the utility model, the campus patrol robot has a simple structure, a small size, and a light weight. It is suitable for environments with a large number of people, such as campuses. It uses crawler tracks to walk, has good stability, can travel on complex terrains, can replace humans to work in dangerous or harmful environments, and can operate for a long time.

[0034] (2) In the present invention, the campus patrol robot uses laser radar and depth camera to collect road conditions and location information. The depth camera is used to obtain the road condition picture in real time, and transmit the collected image data to the Raspberry Pi through the wifi module, and construct the map together with the information collected by the laser radar. The visual SLAM system creates an environmental map and identifies the location and shape of obstacles so that the robot can avoid these obstacles. At the same time, the visual SLAM system can continuously update the map and obstacle information when the robot moves, so that the robot can dynamically avoid new obstacles or environmental changes;

[0035] (3) In the present invention, the campus patrol robot uses an infrared temperature measurement dot matrix sensor and a gas sensor to collect temperature and humidity information and air gas composition information. When the patrol robot detects an abnormal situation, the system automatically activates a voice alarm and reports the data to the staff. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 It is a structural schematic diagram of an automated campus patrol robot based on visual SLAM in the utility model.

[0037] Figure 2 This is a top view of an automated campus patrol robot based on visual SLAM in the utility model.

[0038] Figure 3 It is a structural schematic diagram of the bottom of an automated campus patrol robot based on visual SLAM in the utility model.

[0039] Figure 4 It is a structural schematic diagram of a crawler walking mechanism of an automated campus patrol robot based on visual SLAM in the utility model.

[0040] Figure 5 It is a structural schematic diagram of the top of an automated campus patrol robot based on visual SLAM in the utility model.

[0041] Figure 6 It is a structural schematic diagram of a layered bracket of an automated campus patrol robot based on visual SLAM in the utility model.

[0042] Figure 7 This is a structural program design block diagram of an automated campus patrol robot based on visual SLAM in the utility model.

[0043] In the attached figure: 1. walking chassis; 2. crawler walking mechanism; 3. laser radar; 4. layered bracket; 5. data processing device; 7. depth camera; 8. infrared temperature measurement dot matrix sensor; 9. gas sensor; 21. DC motor; 22. crawler; 23. crawler bracket; 24. driving wheel; 25. driven wheel; 26. tensioning mechanism; 27. tensioning wheel; 41. bottom plate; 42. middle plate; 43. top plate; 44. radar mounting bracket; 45. first leg; 46. second leg. DETAILED DESCRIPTION

[0044] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments, but they are not intended to limit the present invention.

[0045] Please refer to Figures 1 to 7 As shown, an automated campus patrol robot based on visual SLAM is shown, which includes:

[0046] A crawler traveling mechanism 2 and a traveling chassis 1, wherein two crawler traveling mechanisms 2 are symmetrically installed on the left and right sides of the traveling chassis 1;

[0047] A layered support 4, wherein the layered support 4 is installed on the top of the walking chassis 1;

[0048] The gas sensor 9 and the infrared temperature measurement dot matrix sensor 8 are both mounted on the layered bracket 4;

[0049] Laser radar 3, the laser radar 3 is installed on the top of the layered bracket 4;

[0050] A single servo and a depth camera 7, wherein the single servo is mounted on a layered bracket 4, and the depth camera 7 is mounted on the single servo;

[0051] The data processing device 5 is installed in the layered bracket 4, and the laser radar 3, the gas sensor 9, the infrared temperature measurement dot matrix sensor 8, the driving servo and the depth camera 7 are all connected to the data processing device by signal.

[0052] Furthermore, in a preferred embodiment, the layered bracket 4 includes: a top plate 43, a middle plate 42, a bottom plate 41, a first leg 45 and a second leg 46. The lower surface of the bottom plate 41 is connected to the top of the walking chassis 1, a second leg 46 is installed at each corner of the middle plate 42, and the lower ends of multiple second legs 46 are connected to the upper surface of the bottom plate 41; a first leg 45 is installed at each corner of the lower surface of the top plate 43, and the lower ends of multiple first legs 45 are connected to the upper surface of the middle plate 42.

[0053] Furthermore, in a preferred embodiment, the layered bracket 4 also includes: a radar mounting bracket 44 , the radar mounting bracket 44 is installed on the upper surface of the top plate 43 , and the laser radar 3 is installed in the radar mounting bracket 44 .

[0054] Furthermore, in a preferred embodiment, the data processing device 5 is installed on the middle board 42 .

[0055] Further, in a preferred embodiment, the crawler walking mechanism 2 includes: a crawler bracket 23, a driving wheel 24, a driven wheel 25, a tensioning mechanism 26, a tensioning wheel 27 and a crawler 22, the crawler bracket 23 is installed on the walking chassis 1, and a plurality of driven wheels 25 are rotatably installed at the bottom of the crawler bracket 23, the tensioning mechanism 26 is installed on the front side of the crawler bracket 23, the driving wheel 24 is installed on the rear side of the crawler bracket, the tensioning wheel 27 is rotatably installed on the tensioning mechanism 26, and the crawler 22 is arranged around the driving wheel 24, the tensioning wheel 27 and a plurality of driven wheels 25.

[0056] Furthermore, in a preferred embodiment, the crawler walking mechanism 2 further includes: a DC motor 21 , the DC motor 21 is mounted on a crawler bracket 23 , and a driving wheel 24 is mounted at an output end of the DC motor 21 .

[0057] Furthermore, in a preferred embodiment, the crawler walking mechanism 2 also includes: a motor driver and a controller, the motor driver is installed on the walking chassis 1, the controller is installed on the layered bracket 4, the motor driver and the controller are connected through lines, and the motor driver and the DC motor 21 are connected through lines.

[0058] Furthermore, in a preferred embodiment, it also includes: a power supply device, and the crawler walking mechanism 2, the laser radar 3, the data processing device 5, the depth camera 7, the infrared temperature measurement dot matrix sensor 8 and the gas sensor 9 are all connected to the power supply device through a circuit.

[0059] Furthermore, in a preferred embodiment, the data processing device includes: a single-chip microcomputer and a Raspberry Pi, the single-chip microcomputer and the Raspberry Pi are both connected to the power supply device through a circuit, and the single-chip microcomputer and the Raspberry Pi are signal-connected.

[0060] Furthermore, in a preferred embodiment, it also includes: a remote control device, the remote control device and the data processing device are connected by wireless signals, and the remote control device is provided with a display screen and an alarm device.

[0061] The above are only preferred embodiments of the present invention, and are not intended to limit the implementation methods and protection scope of the present invention.

[0062] The utility model also has the following implementation methods based on the above:

[0063] In a further embodiment of the utility model, the robot's appearance design is as follows: the patrol robot's wheels are designed with tracks to increase the contact area with the ground, so that it has a shock-absorbing function; the chassis bracket of the vehicle body uses an aluminum alloy as a frame, and has the characteristics of a general crawler walking mechanism "four wheels and one belt", a tensioning device and a buffer spring; the chassis body is made of stainless steel plates, so that the body has sufficient strength and rigidity and is not easy to deform; there are multiple movement modes, including forward, backward, turning and other movement modes, which can effectively adapt to various changing campus environments.

[0064] In a further embodiment of the utility model, the positioning and mapping of the patrol robot adopts a combination of visual SLAM and laser SLAM algorithms. Visual SLAM (Simultaneous Localization and Mapping) is a vision-based simultaneous positioning and mapping algorithm. In visual SLAM, the robot uses a camera as the main sensor, and constructs an environmental map in real time by processing the image information obtained from the camera, and determines its own position in the map. The visual SLAM algorithm is easily affected by the surrounding environment characteristics, light intensity, operation mode, etc. Therefore, after the laser radar 3 and the depth camera 7 collect environmental information, the organic combination of visual SLAM and laser SLAM algorithms can improve positioning accuracy, enhance environmental perception capabilities, and improve stability and reliability.

[0065] In a further embodiment of the present invention, visual SLAM and laser SLAM are combined to provide more accurate pose estimation by utilizing the high resolution of visual information and the high precision distance measurement of the laser radar 3. Laser SLAM can provide stable positioning support when visual information is insufficient, such as in low light or no light environment.

[0066] In a further embodiment of the present invention, visual SLAM provides rich environmental visual information, can identify details such as color and texture in the environment, and help the robot better understand and interact with the environment. Laser SLAM provides precise geometric information to help the robot accurately navigate and avoid obstacles in complex environments.

[0067] In a further embodiment of the utility model, through data fusion, the vision and laser sensors can back up each other and enhance the stability of the system. When one sensor fails due to environmental changes, such as changes in lighting, the other sensor can continue to provide reliable data. The data fusion algorithm can comprehensively process the vision and laser data, eliminate the errors caused by a single sensor, and improve the stability of the overall system.

[0068] In a further embodiment of the utility model, according to the characteristics of the campus environment, specific sensors are selected for information recognition, such as infrared temperature measurement dot matrix sensor 8 and gas sensor 9, which detect the temperature and gas composition in the environment to determine the occurrence of special abnormal phenomena in the campus environment and respond and alarm in time.

[0069] In a further embodiment of the present invention, for the temperature detection threshold: the infrared temperature measurement dot matrix sensor 8 is used to monitor the ambient temperature and human body temperature. When the ambient temperature exceeds the set upper threshold or is lower than the set lower threshold, the system will trigger an alarm. The specific threshold range is as follows: Upper threshold: 60°C. If the ambient temperature is higher than 60°C, it may indicate a fire or overheating of the equipment. Lower threshold: 0°C. If the ambient temperature is lower than 0°C, it may indicate that the equipment is frozen or the ambient temperature is abnormal.

[0070] When it detects the human body temperature range. The specific threshold range is as follows: Temperature range: 36℃~38℃ (When patrolling at night, it can determine whether there are people in the laboratory.

[0071] In a further embodiment of the utility model, a gas sensor 9 is used to monitor the concentration of harmful gases, dust and smoke in the environment. When it is detected that the concentration of harmful gases, dust or smoke exceeds the safety standard, the system will trigger an alarm. The specific threshold ranges are as follows: Carbon monoxide (CO): exceeding 35ppm. If the concentration of carbon monoxide in the environment exceeds 35ppm, it indicates that there may be a fire hazard or equipment leakage. Carbon dioxide (CO2): exceeding 1000ppm. If the concentration of carbon dioxide in the environment exceeds 1000ppm, it indicates poor ventilation or crowded people. Dust concentration: exceeding 0.1mg / m 3 If the dust concentration in the environment exceeds 0.1 mg / m 3 , which may be harmful to health. Smoke density: The presence of smoke is detected. If smoke is detected, it means there may be a fire.

[0072] In a further embodiment of the utility model, when it is detected that the temperature or gas composition exceeds the set threshold, the patrol robot will trigger a voice alarm, and upload the data to the manager's mobile device through the wireless communication module, notifying the manager to take corresponding measures. The processing flow is: First, data collection: the sensor collects ambient temperature and gas composition data in real time. Secondly, data processing: the microcontroller and Raspberry Pie analyze and process the collected data to determine whether it exceeds the threshold. Then the alarm is triggered: if the data exceeds the threshold, the system will automatically trigger a voice alarm and upload the abnormal data to the management platform. Finally, the manager responds: after receiving the alarm notification, the manager takes timely measures to deal with the abnormal situation to ensure the safety of the campus environment.

[0073] In a further embodiment of the utility model, the campus patrol robot has significant advantages in monitoring the computer room. It can move freely to any monitoring area according to actual needs, realize all-round monitoring of the computer room, and ensure that the environment is safe and free of abnormalities. Image data is obtained through a depth camera, and feature points in the image, such as corners, edges or other areas with unique textures, are extracted based on the SIFT feature extraction algorithm. At the same time, environmental data information is obtained through a laser radar. After feature matching is performed in the Raspberry Pi, the PnP problem solving algorithm is used to perform posture estimation. The map is constructed based on the data obtained above, and then the accuracy of the map and positioning is improved through optimization algorithms, such as graph optimization (Graph Optimization) and bundle adjustment (BundleAdjustment).

[0074] In a further embodiment of the utility model, the configuration and task setting of the campus patrol robot system are also very flexible, and the patrol tasks can be freely configured according to the requirements of time, content and path. The system software design allows users to send signals to the Raspberry Pi or the single-chip microcomputer through the display screen to configure the threshold of the environmental parameters, and can also modify the alarm conditions in real time. According to different conditions, such as night and day patrols, different thresholds of environmental parameters (such as temperature) are configured for detection. For example, when patrolling different types of laboratories, the thresholds of gas parameters such as carbon monoxide and carbon dioxide can be adjusted to detect the environment of the laboratory.

[0075] In a further embodiment of the utility model, the built-in alarm system maximizes the safety of the campus environment, and can promptly alarm and notify management personnel in emergency situations to ensure the safety of the environment and personnel. When it is detected that the environmental parameters exceed the set threshold, the system will immediately trigger an alarm and notify management personnel to handle it in time. The alarm methods include voice alarm and wireless communication. The function of data recording and tracing, the system automatically records the environmental data and alarm events of each detection, and provides detailed log records to facilitate subsequent analysis and tracing; managers can analyze the causes and frequencies of abnormal events through data records and formulate more effective preventive measures. Remote monitoring and management, managers can view the status and environmental data of the robot in real time through the remote control platform, perform remote operation and scheduling, set multiple security policies, ensure the safe transmission and storage of data, and prevent unauthorized access and operation.

[0076] In a further embodiment of the utility model, the automatic campus patrol robot based on visual SLAM is composed of multiple modules, including an environment detection module, a mapping module, an obstacle avoidance module, a data processing module and an alarm module. The system monitors the campus indoor environment through these modules.

[0077] In a further embodiment of the utility model, the infrared temperature measurement dot matrix sensor 8 is used to monitor the temperature of the indoor environment to avoid the occurrence of indoor fire. The gas sensor 9 is used to detect the concentration of harmful gases, dust, and smoke in the environment, such as abnormal gas phenomena such as carbon monoxide, carbon dioxide, and methane.

[0078] In a further embodiment of the utility model, the laser radar 3 collects environmental information, constructs a map, and uses a visual SLAM algorithm and a laser SLAM algorithm for path planning.

[0079] In a further embodiment of the present invention, the depth camera 7 is used to obtain road condition images in real time. The depth camera 7 transmits the collected image data to the Raspberry Pi through a wifi module, and constructs a map together with the information collected by the laser radar 3.

[0080] In a further embodiment of the utility model, the obstacle avoidance module is used for the fixed-point movement of the patrol robot. When the patrol robot senses static and dynamic objects that hinder its passage through sensors based on the collected state information of the obstacles during walking, it first uses an algorithm and then controls the obstacle avoidance. The visual SLAM system can help solve the obstacle avoidance problem and finally reach the target point.

[0081] In a further embodiment of the present invention, environment modeling: the visual SLAM system identifies the location and shape of obstacles by creating an environment map. These maps can be used for subsequent path planning and obstacle avoidance decisions.

[0082] In a further embodiment of the present invention, obstacle detection: During the process of building a map, the visual SLAM system usually detects and identifies obstacles in the environment. This information can be passed to the obstacle avoidance system so that the robot can avoid these obstacles.

[0083] In a further embodiment of the present invention, path planning: based on the established map and environmental obstacle information, the visual SLAM system can assist the path planning algorithm in selecting a safe path and avoiding known obstacles.

[0084] In a further embodiment of the utility model, real-time update: the visual SLAM system can continuously update the map and obstacle information when the robot moves, so that the robot can dynamically avoid newly appearing obstacles or environmental changes.

[0085] In a further embodiment of the present invention, sensor fusion: the visual SLAM system is usually used in combination with other sensors to provide a more comprehensive environmental perception. This fusion can further enhance the robot's obstacle avoidance capabilities and improve the reliability and accuracy of the system through cross-validation of multi-sensor data.

[0086] In a further embodiment of the utility model, the visual SLAM system provides important support for the robot's autonomous navigation and obstacle avoidance decision-making by providing environmental maps and obstacle detection information. However, direct obstacle avoidance functions usually need to be implemented through other specialized obstacle avoidance algorithms or systems, and the visual SLAM system provides the necessary data basis for these algorithms.

[0087] In a further embodiment of the utility model, the data processing device 5 includes: 32 single-chip microcomputers (STM32) and Raspberry Pi. The single-chip microcomputer STM32 drive-control integrated device performs environmental detection and motion control through sensors such as infrared temperature measurement dot matrix sensors 8, gas sensors 9 and motor encoders to realize the judgment of emergencies. The Raspberry Pi completes path planning by processing the information obtained by the laser radar 3 and the depth camera 7.

[0088] In a further embodiment of the utility model, the alarm module adopts DY-SV17F voice broadcast: the voice module will send out a voice alarm signal when the temperature, humidity and air gas composition of the patrol robot exceed the upper limit, the patrol robot enters a blind spot, etc.

[0089] In a further embodiment of the utility model, the system software of the remote control device is designed to respond to abnormal situations. When the patrol robot detects an abnormal situation, the system automatically activates a voice alarm and reports the data to the staff.

[0090] In a further embodiment of the present invention, the data processing and transmission process:

[0091] Data acquisition: Infrared temperature measurement dot matrix sensor 8 and gas sensor 9 detect the environment to obtain environmental data; laser radar 3 and depth camera 7 capture position information in real time and transmit it to the host computer through the serial port for processing.

[0092] Data processing: The 32-bit single-chip computer detects the environment and controls the motion through infrared temperature measurement dot matrix sensors, gas sensors 9 and other sensors and motor encoders to judge emergencies, and execute corresponding operation instructions, such as voice alarm and uploading abnormal data to the mobile terminal of the management personnel.

[0093] Host computer Raspberry Pi: The information obtained by the laser radar 3 and the depth camera 7 is processed based on the visual SLAM and laser SLAM algorithms to complete the path planning.

[0094] In a further embodiment of the utility model, the 32 single-chip microcomputer and the host computer Raspberry Pi exchange information through the serial port, thereby completing the real-time monitoring of the vehicle body and the campus environment.

[0095] In a further embodiment of the utility model, the user can configure the robot's status through the control system and receive information detected by the robot. When the system detects an abnormal situation, the data processing module will automatically sound a voice alarm, upload information to the software control system, and display it on the robot display screen. The alarm can only be released after the administrator enters the correct password.

[0096] In a further embodiment of the utility model, the two crawler walking mechanisms 2 are driven by two DC motors 21 respectively. By controlling the rotation speed and opening and closing of the two DC motors 21 respectively, the crawler walking mechanisms 2 on both sides can be driven separately to realize forward, backward and turning movement modes.

[0097] In a further embodiment of the utility model, the DC motor 21, the motor driver and the controller are connected through lines, the controller is connected to the data processing device 5, and the controller is controlled by the data processing device 5 to realize the automatic walking function of the campus patrol robot.

[0098] In a further embodiment of the utility model, the infrared temperature measurement dot matrix sensor 8 is used to monitor the temperature of the environment, and the data processing device 5 collects the data and transmits it to the remote control device. When the ambient temperature exceeds the threshold, the alarm device on the remote control device emits an alarm sound and sounds an alarm.

[0099] In a further embodiment of the utility model, the gas sensor 9 is used to detect abnormal gas phenomena such as harmful gases, dust, smoke concentration, etc. in the environment. The data processing device 5 collects data and transmits it to the remote control device. When the gas concentration exceeds the threshold, the alarm device on the remote control device emits an alarm sound and alarms.

[0100] In a further embodiment of the utility model, the laser radar 3 collects environmental information to obtain position information, the depth camera 7 is used to obtain road conditions in real time, the data processing device 5 collects data and transmits it to the remote control device, and the Raspberry Pi processes the information obtained by the laser radar 3 and the depth camera 7 to complete path planning and control the crawler walking mechanism 2 to walk according to the planned path.

[0101] In a further embodiment of the utility model, the single-chip microcomputer is used to collect and transmit information, and a wifi module is installed on it. The single-chip microcomputer and the remote control device are remotely connected through wireless signals to realize signal transmission. The single-chip microcomputer and the Raspberry Pi exchange information through the serial port to realize signal transmission and control.

[0102] In a further embodiment of the utility model, the depth camera 7 is used to collect road condition information and identify obstacles. After the collected information is transmitted to the Raspberry Pi, the Raspberry Pi replans the path to facilitate subsequent walking and realize the obstacle avoidance function.

[0103] In a further embodiment of the utility model, the laser radar 3 is located at the highest position of the campus patrol robot to ensure the accuracy of positioning.

[0104] In a further embodiment of the present invention, the depth camera 7 is controlled by a single servo, can rotate left and right, and is located on the top front side of the walking chassis to prevent the depth camera 7 from being blocked by lines and other structural parts.

[0105] In a further embodiment of the utility model, a buffer spring is provided inside the tensioning mechanism 26. By providing the buffer spring to push the tensioning wheel 27, the crawler 22 is in a tensioned state, ensuring that the crawler walking mechanism 2 can walk stably.

[0106] In a further embodiment of the present invention, a supporting wheel is installed on the crawler support 24 , and the supporting wheel is arranged above the tensioning mechanism 26 to support the top crawler 22 .

[0107] The above are only preferred embodiments of the present invention, and do not limit the implementation mode and protection scope of the present invention. Those skilled in the art should be aware that all solutions obtained by equivalent substitutions and obvious changes made using the description and illustrations of the present invention should be included in the protection scope of the present invention.

Claims

1. An automated campus patrol robot based on visual SLAM, characterized in that: include: A crawler walking mechanism (2) and a walking chassis (1), wherein the two crawler walking mechanisms (2) are symmetrically mounted on the left and right sides of the walking chassis (1); A layered support (4), the layered support (4) is installed on the top of the walking chassis (1); A gas sensor (9) and an infrared temperature measurement dot matrix sensor (8), wherein the gas sensor (9) and the infrared temperature measurement dot matrix sensor (8) are both mounted on a layered support (4); A laser radar (3), wherein the laser radar (3) is mounted on the top of the layered support (4); A single steering gear and a depth camera (7), wherein the single steering gear is mounted on a layered bracket (4), and the depth camera (7) is mounted on the single steering gear; A data processing device (5) is installed in the layered support (4); the laser radar (3), the gas sensor (9), the infrared temperature measurement dot matrix sensor (8), the driving servo and the depth camera (7) are all connected to the data processing device by signals.

2. The automated campus patrol robot based on visual SLAM according to claim 1, characterized in that: The layered support (4) comprises: a top plate (43), a middle plate (42), a bottom plate (41), a first leg (45) and a second leg (46); the lower surface of the bottom plate (41) is connected to the top of the walking chassis (1); a second leg (46) is installed at each corner of the middle plate (42); the lower ends of the plurality of second legs (46) are connected to the upper surface of the bottom plate (41); a first leg (45) is installed at each corner of the lower surface of the top plate (43); the lower ends of the plurality of first legs (45) are connected to the upper surface of the middle plate (42).

3. The automated campus patrol robot based on visual SLAM according to claim 2, characterized in that: The layered support (4) further comprises: a radar mounting support (44), wherein the radar mounting support (44) is mounted on the upper surface of the top layer board (43), and the laser radar (3) is mounted in the radar mounting support (44).

4. The automated campus patrol robot based on visual SLAM according to claim 2, characterized in that: The data processing device (5) is installed on the middle board (42).

5. The automated campus patrol robot based on visual SLAM according to claim 1, characterized in that: The crawler walking mechanism (2) comprises: a crawler support (23), a driving wheel (24), a driven wheel (25), a tensioning mechanism (26), a tensioning wheel (27) and a crawler track (22); the crawler support (23) is mounted on the walking chassis (1); a plurality of driven wheels (25) are rotatably mounted on the bottom of the crawler support (23); the tensioning mechanism (26) is mounted on the front side of the crawler support (23); the driving wheel (24) is mounted on the rear side of the crawler support; the tensioning wheel (27) is rotatably mounted on the tensioning mechanism (26); and the crawler track (22) is arranged around the driving wheel (24), the tensioning wheel (27) and the plurality of driven wheels (25).

6. The automated campus patrol robot based on visual SLAM according to claim 5, characterized in that: The crawler walking mechanism (2) further comprises: a DC motor (21), wherein the DC motor (21) is mounted on a crawler support (23), and the driving wheel (24) is mounted on the output end of the DC motor (21).

7. The automated campus patrol robot based on visual SLAM according to claim 6, characterized in that: The crawler walking mechanism (2) also includes: a motor driver and a controller, the motor driver is installed on the walking chassis (1), the controller is installed on the layered bracket (4), the motor driver and the controller are connected via a line, and the motor driver and the DC motor (21) are connected via a line.

8. The automated campus patrol robot based on visual SLAM according to claim 1, characterized in that: Also includes: A power supply device, the crawler walking mechanism (2), the laser radar (3), the data processing device (5), the depth camera (7), the infrared temperature measurement dot matrix sensor (8) and the gas sensor (9) are all connected to the power supply device through a circuit.

9. The automated campus patrol robot based on visual SLAM according to claim 8, characterized in that: The data processing device comprises: a single chip microcomputer and a Raspberry Pi, the single chip microcomputer and the Raspberry Pi are both connected to a power supply device through a circuit, and the single chip microcomputer and the Raspberry Pi are signal connected.

10. The automated campus patrol robot based on visual SLAM according to claim 1, characterized in that: It also includes: a remote control device, the remote control device and the data processing device are connected by wireless signals, and the remote control device is provided with a display screen and an alarm device.

Images