ROS construction site inspection robot device based on multi-dimensional sensor array integration and autonomous navigation optimization algorithm

Abstract

The invention discloses an ROS construction site inspection robot device based on multi-dimensional sensor array integration and an autonomous navigation optimization algorithm. The ROS construction site inspection robot device comprises a mechanical vehicle body, a driving module, a multi-dimensional sensor detection module, a signal transmission module and an upper computer. According to the robot device, a multi-dimensional sensor is adopted for safety detection, an SLAM navigation optimization algorithm is fused to achieve autonomous navigation, data collection is achieved through integration of the Bluetooth technology and a rear-end platform, and the robot device aims at optimizing the efficiency in construction site inspection work, improving the personnel safety in the work process and improving the working efficiency. And the real-time performance and reliability of data collection are ensured. Various advanced technologies such as SLAM, target detection and recognition, autonomous navigation and obstacle avoidance, real-time data acquisition and processing and the like are integrated, and great potential is shown in the aspect of improving inspection efficiency, safety and accuracy.

Description

Technical Field

[0001] The present invention relates to the field of robots, and more particularly to a ROS construction site inspection robot device based on multi-dimensional sensor array integration and autonomous navigation optimization algorithm. Background Art

[0002] In recent years, the construction industry has faced multiple challenges such as safety hazards, labor shortages, and rising costs. The development of intelligent construction site inspection robots has emerged as an important means to improve construction efficiency and ensure safety. First, the work of site inspection is heavy and time-consuming. Traditional manual inspections are not only inefficient but also easily affected by factors such as weather and worker fatigue, leading to an increase in safety hazards. Second, the construction site environment is complex and there are many potential dangers, such as working at heights and heavy object falls. Intelligent robots can use sensors and intelligent algorithms to monitor and identify dangers in real time, improving safety. Finally, with the digital transformation of the construction industry, more and more enterprises are beginning to pay attention to data collection and analysis. Inspection robots can collect and analyze data in real time to provide support for decision-making.

[0003] The environments, equipment layouts, and inspection requirements of different construction sites vary, which requires inspection robots to have high adaptability and flexibility. However, the current inspection robots on the market may not fully meet the needs of all construction sites. Especially in complex and changing construction site environments, their applications are restricted to a certain extent. Existing construction site inspection robots have several disadvantages: 1. When navigating and positioning in complex and changing construction site environments, inspection robots may be affected by obstacles and light changes, resulting in a decrease in positioning accuracy; 2. Some inspection robots may not be equipped with all necessary sensors, or the detection accuracy and range of sensors are limited, resulting in the inability to comprehensively and accurately detect the operating status and potential problems of equipment; 3. It is inevitable that robots will encounter difficulties during navigation. Existing construction site inspection robots cannot respond in a timely and effective manner to possible difficulties, such as driving into a dead end and being in a sensor measurement blind area; 4. In complex and changing construction site environments, the stability of inspection robots may be affected. For example, when driving on uneven ground, the robot may experience bumps and tilts, affecting the accuracy and efficiency of inspection. Summary of the Invention

[0004] The object of the present invention is to provide a ROS construction site inspection robot device based on multi-dimensional sensor array integration and autonomous navigation optimization algorithm, including a mechanical vehicle body, a driving module, a multi-dimensional sensor detection module, a signal transmission module, and a host computer.

[0005] The driving module, the multi-dimensional sensor detection module, and the signal transmission module are mounted on the mechanical vehicle body.

[0006] The driving module drives the mechanical vehicle body to move according to the signals fed back by the host computer.

[0007] The multi-dimensional sensor detection module is used to obtain information about the open-air construction site environment.

[0008] The information about the open-air construction site environment includes at least environmental images.

[0009] The signal transmission module is used to transmit the information about the open-air construction site environment obtained by the multi-dimensional sensor detection module to the host computer.

[0010] The host computer constructs a construction site environment map based on the information about the open-air construction site environment, plans the moving path of the ROS construction site inspection robot device based on the autonomous navigation optimization algorithm, and generates a feedback signal.

[0011] The host computer transmits the feedback signal to the driving module.

[0012] Furthermore, the mechanical vehicle body includes a wheeled mechanical vehicle.

[0013] Furthermore, the material of the mechanical vehicle body includes carbon fiber composite material.

[0014] Furthermore, the driving module includes a power source and a motion mechanism.

[0015] The power source provides power for the motion mechanism.

[0016] The power source includes two-way motor drive modules and an energy storage battery.

[0017] The energy storage battery supplies energy to the two-way motor drive modules.

[0018] One-way motor drive module is used to drive the movement of the mechanical vehicle body, and the other one-way motor drive module is used to drive the motion mechanism.

[0019] The motion mechanism is used to drive the mechanical vehicle body to rotate.

[0020] The motion mechanism includes a Mecanum omnidirectional wheel mechanism.

[0021] Furthermore, the multi-dimensional sensor detection module includes a distance sensor, an inertial sensor, a vision sensor, and a temperature and humidity sensor.

[0022] The distance sensor is used to detect the distance between the obstacle and the ROS construction site inspection robot device and the distance between the ROS construction site inspection robot device and the target position.

[0023] The target position is the end point of the moving path of the ROS construction site inspection robot device.

[0024] The inertial sensor is used to monitor the motion state of the inspection robot device in real time and provide pose information.

[0025] The vision sensor is used to obtain images of the surrounding environment of the ROS construction site inspection robot device.

[0026] The temperature and humidity sensor is used to detect the environmental temperature and humidity.

[0027] Furthermore, the signal transmission methods of the signal transmission module include Bluetooth transmission and serial port transmission.

[0028] Furthermore, the steps for the host computer to construct a construction site environment map based on the information of the open-air construction site environment are as follows:

[0029] s11 Obtain the information of the open-air construction site environment through the multi-dimensional sensor detection module.

[0030] s12 Preprocess the obtained information of the open-air construction site environment to obtain preprocessed data.

[0031] The preprocessing includes noise removal, smoothing processing, and feature extraction.

[0032] s13 Perform particle filtering on the preprocessed data based on the Monte Carlo method to evaluate the position of the ROS construction site inspection robot device in the environment.

[0033] s14 Construct a construction site environment map according to the evaluated position and the information of the open-air construction site environment.

[0034] Furthermore, the steps for the host computer to plan the moving path of the ROS construction site inspection robot device based on the autonomous navigation optimization algorithm and generate a feedback signal are as follows:

[0035] s21 Set the initial position and the target position of the ROS construction site inspection robot device.

[0036] The target position is the end point of the moving path of the ROS construction site inspection robot device.

[0037] s22 According to the information of the open-air construction site environment obtained by the multi-dimensional sensor detection module, construct a construction site environment map in real time and evaluate the position of the ROS construction site inspection robot device in the environment.

[0038] s23 Determine whether there are obstacles in the surrounding environment of the ROS construction site inspection robot device. If so, use the Cosmap cost map to avoid obstacles. If not, go to step s24.

[0039] The steps for using the Cosmap cost map to avoid obstacles are as follows:

[0040] s231 obtains the conversion relationship between the global coordinate system and the robot coordinate system and loads the layers.

[0041] The layers include a static map layer, an obstacle map layer, and an inflation layer.

[0042] s232 updates the obstacle map layer in real time based on the information of the open-air construction site environment and the motion state of the ROS construction site inspection robot device, and calculates the inflation layer based on the static map layer and the obstacle map layer.

[0043] s233 transmits the updated data to step s24.

[0044] s24 calculates the optimal path from the current position to the target position according to the real-time constructed construction site environment map and the position of the ROS construction site inspection robot device in the environment, combines the global path planner and the local path planner, and generates a feedback signal to drive the movement of the vehicle body, and returns to step s22 until the ROS construction site inspection robot device reaches the target position.

[0045] The global path planner includes A ＊ algorithm.

[0046] The local path planner includes the DWA algorithm.

[0047] Furthermore, the ROS construction site inspection robot device is also provided with an error prompt module and a recovery behavior module.

[0048] In the absence of dynamic obstacles, if the ROS construction site inspection robot device cannot reach the target position, the error prompt module sends an error signal to the user.

[0049] When the host computer detects that the ROS construction site inspection robot device is stuck by an obstacle and cannot move forward, the recovery behavior module executes the recovery behavior.

[0050] Furthermore, the ROS construction site inspection robot device also includes a human-machine interaction interface.

[0051] The human-machine interaction interface is used to display the information of the open-air construction site environment and the construction site environment map.

[0052] The technical effects of the present invention are beyond doubt. The present invention integrates a variety of advanced technologies such as SLAM, target detection and recognition, autonomous navigation and obstacle avoidance, real-time data acquisition and processing, etc., and shows great potential in improving the inspection efficiency, safety and accuracy.

[0053] The present invention provides a robot device that uses multi-dimensional sensors for safety detection, integrates SLAM navigation optimization algorithms to achieve autonomous navigation, and integrates with a backend platform through Bluetooth technology to collect data. The device aims to optimize the efficiency in construction site inspections, improve the safety of personnel during the operation process, and ensure the real-time and reliability of data collection.

[0054] The beneficial effects of the present invention include:

[0055] 1) It can achieve high-precision positioning in a complex construction site environment. The present invention adopts SLAM positioning technology, continuously fuses the sensor data of lidar and camera, and the robot motion information, and is equipped with algorithms such as gmapping, cartographer, and hector_slam, which has high robustness and can update and optimize the robot's position estimation in real time in an unknown environment and complete map construction.

[0056] 2) It is equipped with a variety of sensors and can comprehensively collect surrounding environment data. The present invention adopts the lidar of SlAM A1, RGBD depth camera, 6-axis IMU sensor and temperature and humidity controller, which can automatically and continuously monitor various parameters of the construction site, improve the inspection efficiency and enhance the inspection accuracy.

[0057] 3) The present invention provides a recovery strategy for the global cost map and local cost map, and integrates recovery strategy plugins such as rotate_recovery, move_slow_and_clear, and clear_costmap_recovery. When the robot feels stuck, the navigation node will choose to execute recovery behaviors, such as rotating in place to forcefully clear the residual obstacles in the cost map or moving forward or backward at a very small speed to get out of trouble.

[0058] 4) The present invention adopts a pendulum-mounted Mecanum wheel, which can achieve omnidirectional movement, effectively avoid slipping, and can adapt to complex and changeable environments. Description of the Drawings

[0059] Figure 1 It is a schematic diagram of the basic structure of the mechanism device;

[0060] Figure 2 It is a schematic diagram of the structure of the pendulum-mounted Mecanum wheel;

[0061] Figure 3 It is a flow chart of map building with the Gmapping algorithm as an example;

[0062] Figure 4 It is a schematic diagram of the path planning process; Figure 4 (a) It is a schematic diagram of the path planning process without dynamic obstacles; Figure 4(b) Schematic diagram of the path planning process for a single dynamic obstacle; Figure 4 (c) Schematic diagram of the path planning process for multiple dynamic obstacles;

[0063] Figure 5 Schematic diagram of the process of using the Costmap cost map method to draw the obstacle description layer under complex obstacles;

[0064] Figure 6 Robot recovery process with rotate_recovery as a plugin;

[0065] Figure 7 Basic process of the temperature and humidity sensor to achieve Bluetooth signal transmission;

[0066] Figure 8 Schematic diagram of the system structure of the present invention. Detailed implementation manners

[0067] The present invention will be further described below in conjunction with embodiments, but it should not be understood that the above-mentioned subject matter scope of the present invention is limited to the following embodiments. Without departing from the above-mentioned technical idea of the present invention, various substitutions and changes made according to the common general knowledge and customary means in the art shall all be included within the protection scope of the present invention.

[0068] Embodiment 1:

[0069] Refer to Figures 1 to 8 , a ROS construction site inspection robot device based on multi-dimensional sensor array integration and autonomous navigation optimization algorithm, including a mechanical vehicle body, a driving module, a multi-dimensional sensor detection module, a signal transmission module, and a host computer.

[0070] The driving module, the multi-dimensional sensor detection module, and the signal transmission module are mounted on the mechanical vehicle body.

[0071] The driving module drives the mechanical vehicle body to move according to the signal feedback from the host computer.

[0072] The multi-dimensional sensor detection module is used to obtain information about the open construction site environment.

[0073] The information about the open construction site environment at least includes environmental images.

[0074] The signal transmission module is used to transmit the information about the open construction site environment obtained by the multi-dimensional sensor detection module to the host computer.

[0075] The host computer constructs a construction site environment map based on the information about the open construction site environment, and plans the movement path of the ROS construction site inspection robot device based on the autonomous navigation optimization algorithm, and generates a feedback signal.

[0076] The host computer transmits the feedback signal to the drive module.

[0077] Embodiment 2:

[0078] A ROS construction site inspection robot device based on multi-dimensional sensor array integration and autonomous navigation optimization algorithm. The main technical content is shown in Embodiment 1. Further, the mechanical vehicle body includes a wheeled mechanical vehicle.

[0079] Embodiment 3:

[0080] A ROS construction site inspection robot device based on multi-dimensional sensor array integration and autonomous navigation optimization algorithm. The main technical content is shown in any one of Embodiments 1 to 2. Further, the material of the mechanical vehicle body includes carbon fiber composite material.

[0081] Embodiment 4:

[0082] A ROS construction site inspection robot device based on multi-dimensional sensor array integration and autonomous navigation optimization algorithm. The main technical content is shown in any one of Embodiments 1 to 3. Further, the drive module includes a power source and a motion mechanism.

[0083] The power source provides power for the motion mechanism.

[0084] The power source includes two-way motor drive modules and an energy storage battery.

[0085] The energy storage battery supplies energy to the two-way motor drive modules.

[0086] One-way motor drive module is used to drive the movement of the mechanical vehicle body, and the other one-way motor drive module is used to drive the motion mechanism.

[0087] The motion mechanism is used to drive the mechanical vehicle body to rotate.

[0088] The motion mechanism includes a Mecanum omnidirectional wheel mechanism.

[0089] Embodiment 5:

[0090] A ROS construction site inspection robot device based on multi-dimensional sensor array integration and autonomous navigation optimization algorithm. The main technical content is shown in any one of Embodiments 1 to 4. Further, the multi-dimensional sensor detection module includes a distance sensor, an inertial sensor, a vision sensor, and a temperature and humidity sensor.

[0091] The distance sensor is used to detect the distance between the obstacle and the ROS construction site inspection robot device and the distance between the ROS construction site inspection robot device and the target position.

[0092] The target position is the end point of the movement path of the ROS construction site inspection robot device.

[0093] The inertial sensor is used to monitor the motion state of the inspection robot device in real time and provide pose information.

[0094] The vision sensor is used to obtain images of the surrounding environment of the ROS construction site inspection robot device.

[0095] The temperature and humidity sensor is used to detect the environmental temperature and humidity.

[0096] Embodiment 6:

[0097] A ROS construction site inspection robot device based on multi-dimensional sensor array integration and autonomous navigation optimization algorithm, the main technical content is as described in any one of Embodiments 1 to 5. Further, the signal transmission methods of the signal transmission module include Bluetooth transmission and serial port transmission.

[0098] Embodiment 7:

[0099] A ROS construction site inspection robot device based on multi-dimensional sensor array integration and autonomous navigation optimization algorithm, the main technical content is as described in any one of Embodiments 1 to 6. Further, the steps for the host computer to construct a construction site environment map according to the information of the open-air construction site environment are as follows:

[0100] s11 Obtain the information of the open-air construction site environment through the multi-dimensional sensor detection module.

[0101] s12 Preprocess the obtained information of the open-air construction site environment to obtain preprocessed data.

[0102] The preprocessing includes noise removal, smoothing processing, and feature extraction.

[0103] s13 Perform particle filtering on the preprocessed data based on the Monte Carlo method to evaluate the position of the ROS construction site inspection robot device in the environment.

[0104] s14 Construct a construction site environment map according to the evaluated position and the information of the open-air construction site environment.

[0105] Embodiment 8:

[0106] A ROS construction site inspection robot device based on multi-dimensional sensor array integration and autonomous navigation optimization algorithm, the main technical content is as described in any one of Embodiments 1 to 7. Further, the steps for the host computer to plan the moving path of the ROS construction site inspection robot device based on the autonomous navigation optimization algorithm and generate a feedback signal are as follows:

[0107] s21 Set the initial position and the target position of the ROS construction site inspection robot device.

[0108] The target position is the end point of the moving path of the ROS construction site inspection robot device.

[0109] s22 Based on the information of the open construction site environment obtained by the multi-dimensional sensor detection module, a construction site environment map is constructed in real time, and the position of the ROS construction site inspection robot device in the environment is evaluated.

[0110] s23 Determine whether there are obstacles in the surrounding environment of the ROS construction site inspection robot device. If so, use the Cosmap cost map to avoid obstacles. If not, proceed to step s24.

[0111] The steps of using the Cosmap cost map to avoid obstacles are as follows:

[0112] s231 Obtain the conversion relationship between the global coordinate system and the robot coordinate system, and load the layers.

[0113] The layers include a static map layer, an obstacle map layer, and an inflation layer.

[0114] s232 Update the obstacle map layer in real time through the information of the open construction site environment and the motion state of the ROS construction site inspection robot device, and calculate the inflation layer based on the static map layer and the obstacle map layer.

[0115] s233 Transmit the updated data to step s24.

[0116] s24 According to the construction site environment map constructed in real time and the position of the ROS construction site inspection robot device in the environment, combine the global path planner and the local path planner to calculate the optimal path from the current position to the target position, and generate a feedback signal to drive the movement of the mechanical vehicle body, and return to step s22 until the ROS construction site inspection robot device reaches the target position.

[0117] The global path planner includes A ＊ algorithm.

[0118] The local path planner includes the DWA algorithm.

[0119] Embodiment 9:

[0120] A ROS construction site inspection robot device based on multi-dimensional sensor array integration and autonomous navigation optimization algorithm, the main technical content is shown in any one of Embodiments 1 to 8. Further, the ROS construction site inspection robot device is also provided with an error prompt module and a recovery behavior module.

[0121] In the absence of dynamic obstacles, if the ROS construction site inspection robot device cannot reach the target position, the error prompt module sends an error signal to the user.

[0122] When the upper computer detects that the ROS construction site inspection robot device is stuck by an obstacle and cannot move forward, the recovery behavior module executes the recovery behavior.

[0123] Example 10:

[0124] A ROS construction site inspection robot device based on multi - dimensional sensor array integration and autonomous navigation optimization algorithm. The main technical content can be seen in any one of Examples 1 to 9. Further, the ROS construction site inspection robot device also includes a human - machine interaction interface.

[0125] The human - machine interaction interface is used to display information about the open - air construction site environment and the construction site environment map.

[0126] Example 11:

[0127] See Figures 1 to 8 , a ROS construction site inspection robot device based on multi - dimensional sensor array integration and autonomous navigation optimization algorithm, includes a mechanical vehicle body, a driving module, a multi - dimensional sensor detection module, a signal transmission module, and a host computer.

[0128] The driving module, the multi - dimensional sensor detection module, and the signal transmission module are mounted on the mechanical vehicle body.

[0129] The driving module drives the mechanical vehicle body to move according to the signal fed back by the host computer.

[0130] The multi - dimensional sensor detection module is used to obtain information about the open - air construction site environment.

[0131] In a construction site environment where the ground is underground or the GPS signal cannot be covered, it is indeed challenging to perform precise positioning and management. Therefore, this embodiment is mainly applied to open - air or non - fully enclosed construction sites.

[0132] The information about the open - air construction site environment includes position information, speed information, environmental images, environmental temperature, and environmental humidity.

[0133] The signal transmission module is used to transmit the information about the open - air construction site environment obtained by the multi - dimensional sensor detection module to the host computer.

[0134] The host computer constructs a construction site environment map based on the information about the open - air construction site environment, and plans the moving path of the ROS construction site inspection robot device based on the autonomous navigation optimization algorithm, generating a feedback signal.

[0135] The host computer transmits the feedback signal to the driving module.

[0136] The ROS construction site inspection robot device can perform inspections according to a preset route or flexibly adjust the route according to real - time situations, ensuring the comprehensiveness and timeliness of the inspections.

[0137] In some dangerous, harmful or inaccessible environments, such as high-temperature, high-pressure, toxic, radiation areas, etc., manual inspection poses great safety risks. Robot inspection can avoid direct exposure of personnel to these dangerous environments and effectively ensure the safety of personnel.

[0138] The device carried by the inspection robot can scan and construct the surrounding environment, and can remotely return images to the technicians in the background to monitor the construction process, reducing the limitations of manual inspection. The inspection objects mainly include building entities, construction site personnel and dangerous points.

[0139] The ROS construction site inspection robot device returns data such as whether construction site personnel wear safety helmets and safety vests as required, whether construction site personnel wear masks as required, whether there are personnel exposed in the building danger area, whether the building is collapsed, whether there are high-temperature, high-pressure, toxic, radiation areas, etc. to the background, and the background realizes the inspection of the construction site environment through data processing or manual identification.

[0140] Example 12:

[0141] A ROS construction site inspection robot device based on multi-dimensional sensor array integration and autonomous navigation optimization algorithm, the main technical content is shown in Example 11. Further, the mechanical vehicle body includes a wheeled mechanical vehicle.

[0142] Based on the requirements analysis, a preliminary design scheme of the fuselage is proposed, including determining the mechanical vehicle as the device carrier and determining the basic shape as a wheeled mechanical vehicle.

[0143] Example 13:

[0144] A ROS construction site inspection robot device based on multi-dimensional sensor array integration and autonomous navigation optimization algorithm, the main technical content is shown in any one of Examples 11 to 12. Further, the material of the mechanical vehicle body includes carbon fiber composite material.

[0145] Determine the preliminary dimensions according to the design conditions. According to the complexity of the construction site environment, select high-strength, lightweight carbon fiber composite material as the fuselage material to meet the strength and weight requirements.

[0146] Example 14:

[0147] A ROS construction site inspection robot device based on multi-dimensional sensor array integration and autonomous navigation optimization algorithm, the main technical content is shown in any one of Examples 11 to 13. Further, the drive module includes a power source and a motion mechanism.

[0148] The power source provides power for the motion mechanism.

[0149] The power source includes two-way motor drive modules and energy storage batteries.

[0150] The energy storage battery supplies power to two motor drive modules.

[0151] One motor drive module is used to drive the movement of the mechanical vehicle body, and the other motor drive module is used to drive the movement mechanism.

[0152] The movement mechanism is used to drive the mechanical vehicle body to rotate.

[0153] The movement mechanism includes a Mecanum omnidirectional wheel mechanism.

[0154] Power source selection: According to the running speed and load demand, select high-performance two-motor drive and attach a compatible 12V polymer lithium battery energy storage battery.

[0155] Movement mechanism selection: Clarify the running requirements of the chassis in complex environments, including key parameters such as maximum speed, acceleration, obstacle-crossing ability, climbing angle, minimum turning radius, etc., and select a Mecanum omnidirectional wheel mechanism with a pendulum suspension that can achieve multiple movement postures.

[0156] One motor is mainly used to drive the moving mechanism of the mechanical vehicle body, especially the components directly connected to the bearings or wheels, to achieve the overall movement of the mechanical vehicle. One motor is mainly used to drive the specific movement mechanism of the mechanical vehicle, such as a rotating platform, etc., to achieve more complex operation tasks.

[0157] Example 15:

[0158] A ROS construction site inspection robot device based on multi-dimensional sensor array integration and autonomous navigation optimization algorithm. The main technical content can be seen in any one of Examples 11 to 14. Further, the multi-dimensional sensor detection module includes a ranging sensor, an inertial sensor, a vision sensor, and a temperature and humidity sensor.

[0159] The ranging sensor is used to detect the distance between the obstacle and the ROS construction site inspection robot device and the distance between the ROS construction site inspection robot device and the target position.

[0160] The target position is the end point of the moving path of the ROS construction site inspection robot device.

[0161] Ranging sensor: Adopt the Slang RPLIDAR A1 radar. Using the laser triangulation ranging technology it uses and cooperating with its independently developed high-speed vision acquisition and processing mechanism, it can perform more than 8000 ranging actions per second to ensure fast and accurate environmental perception. Its compatible ROS system is convenient for integration with the ROS robot system.

[0162] The inertial sensor is used to real-time monitor the motion state of the inspection robot device and provide pose information.

[0163] Inertial sensor: An IMU sensor with 6 axes is adopted, which consists of three accelerometers and three gyroscopes. Each accelerometer and gyroscope is responsible for measuring the acceleration and angular velocity of an object along one axis in three-dimensional space respectively.

[0164] By combining data from other sensors (such as lidar, cameras, etc.), the inertial sensor can provide continuous pose information for the robot, assist it in achieving autonomous navigation, and flexibly adjust the travel route according to the predetermined path or environmental changes.

[0165] The inertial sensor can monitor the motion state of the robot in real time, including speed, acceleration, and steering, etc., to ensure the stability and safety of the robot during the inspection process.

[0166] The visual sensor is used to obtain images of the surrounding environment of the ROS construction site inspection robot device.

[0167] Visual sensor: A ROS depth camera is adopted. By using it to obtain the distance information of each point in the image from the camera and combining the two-dimensional image coordinates, a three-dimensional space model is constructed to achieve scene modeling and object recognition. It can be easily integrated into the ROS system and use the algorithms and tools of ROS for data processing and decision-making.

[0168] The temperature and humidity sensor is used to detect the environmental temperature and humidity.

[0169] Temperature and humidity sensor: A DHT11 digital temperature and humidity sensor developed based on the 51 development board and equipped with Bluetooth connection function is adopted.

[0170] When the temperature and humidity exceed the preset safe range, the temperature and humidity sensor can trigger an early warning mechanism to remind the management personnel to take measures in time to avoid potential safety hazards and losses.

[0171] The temperature and humidity sensor can detect and report extreme temperature and humidity conditions in the environment in time, thus avoiding damage to the robot and the equipment it carries due to overheating, overcooling, or excessive humidity.

[0172] Example 16:

[0173] A ROS construction site inspection robot device based on multi-dimensional sensor array integration and autonomous navigation optimization algorithm. The main technical content can be seen in any one of Examples 11 to 15. Further, the signal transmission methods of the signal transmission module include Bluetooth transmission and serial port transmission.

[0174] Example 17:

[0175] A ROS construction site inspection robot device based on multi-dimensional sensor array integration and autonomous navigation optimization algorithm. The main technical content can be found in any one of Embodiments 11 to 16. Further, the steps for the host computer to construct a construction site environment map based on the information of the open-air construction site environment are as follows:

[0176] S11 Obtain the information of the open-air construction site environment through the multi-dimensional sensor detection module.

[0177] S12 Preprocess the obtained information of the open-air construction site environment to obtain preprocessed data.

[0178] The preprocessing includes noise removal, smoothing, and feature extraction.

[0179] S13 Perform particle filtering on the preprocessed data based on the Monte Carlo method to evaluate the position of the ROS construction site inspection robot device in the environment.

[0180] Based on the Monte Carlo method, approximate the pose distribution of the robot through a series of random samples (i.e., particles), and then estimate the position of the robot in the environment.

[0181] S14 Construct a construction site environment map according to the evaluated position and the information of the open-air construction site environment.

[0182] Embodiment 18:

[0183] A ROS construction site inspection robot device based on multi-dimensional sensor array integration and autonomous navigation optimization algorithm. The main technical content can be found in any one of Embodiments 11 to 17. Further, the steps for the host computer to plan the movement path of the ROS construction site inspection robot device based on the autonomous navigation optimization algorithm and generate a feedback signal are as follows:

[0184] S21 Set the initial position and target position of the ROS construction site inspection robot device.

[0185] The target position is the end point of the movement path of the ROS construction site inspection robot device.

[0186] Initial position: The flatness and stability of the ground should be considered to avoid the robot from shaking or tipping over during startup. The initial position should have good visibility and sensing conditions, enabling the robot to quickly perceive the surrounding environment and prepare for subsequent inspection tasks.

[0187] Target position: The target position is usually the area that needs to be focused on during the inspection task. The movement ability and path planning ability of the inspection robot should be considered to ensure that the robot can reach the target position according to the preset route or the autonomously planned route.

[0188] s22 Based on the information of the open-air construction site environment obtained by the multi-dimensional sensor detection module, a construction site environment map is constructed in real time, and the position of the ROS construction site inspection robot device in the environment is evaluated.

[0189] The inspection robot trolley is equipped with a wide-angle camera, and the image can cover most of the panorama. In addition to the camera, the inspection robot trolley is equipped with other types of sensors (such as lidar, etc.). These sensors can provide information on various aspects such as the position, speed, and direction of the robot. By fusing this information with the image data captured by the camera, the robot can more accurately determine its position in the environment.

[0190] s23 Determine whether there are obstacles in the environment around the ROS construction site inspection robot device. If so, use the Cosmap cost map to avoid obstacles. If not, go to step s24.

[0191] Cosmap uses multiple independent rasterized layers to independently maintain obstacle information. These layers can be superimposed to achieve a specific obstacle description layer. Each layer has its own obstacle update rules, such as adding obstacles, deleting obstacles, updating the confidence of obstacle points, etc.

[0192] The steps of using the Cosmap cost map to avoid obstacles are as follows:

[0193] s231 Obtain the conversion relationship between the global coordinate system and the robot coordinate system, and load the layers.

[0194] The layers include a static map layer, an obstacle map layer, and an inflation layer.

[0195] s232 Real-time update the obstacle map layer through the information of the open-air construction site environment and the motion state of the ROS construction site inspection robot device, and calculate the inflation layer based on the static map layer and the obstacle map layer.

[0196] s233 Transmit the updated data to step s24.

[0197] Construct a cosmap cost map:

[0198] Initialization: Obtain the conversion relationship between the global coordinate system and the robot coordinate system, and load each layer (such as the static map layer, the obstacle map layer, the inflation layer, etc.).

[0199] Update: Real-time update the obstacle map layer through the robot's sensor data and motion state, and calculate the inflation layer based on the static map layer and the obstacle map layer.

[0200] Publication: Publish the updated Costmap data for use by modules such as the path planner.

[0201] Specific steps:

[0202] Obtain Costmap data: The path planner reads the cost value distribution of the current environment from the Costmap, including free areas, obstacle areas, and unknown areas.

[0203] Analyze the path: Based on the starting position and target position of the robot, the path planner plans an optimal path from the starting point to the end point on the Costmap. During the planning process, high-cost areas (i.e., obstacle areas and inflation layer areas) will be avoided as much as possible.

[0204] Real-time adjustment: During the process of the robot moving along the planned path, the environmental changes are sensed in real time through sensors, and the obstacle map layer in the Costmap is updated. If new obstacles are found or the positions of existing obstacles change, the path planner will re-plan the path to avoid the obstacles.

[0205] Execute obstacle avoidance: The robot adjusts its own motion state, including speed, direction, etc., according to the new path instructions to ensure that it can safely bypass the obstacles and continue to move towards the target position.

[0206] Continuous monitoring: During the entire movement process, the robot continuously monitors the Costmap data and sensor data to promptly detect and respond to possible new obstacles or changes.

[0207] s24 calculates the optimal path from the current position to the target position based on the real-time constructed construction site environment map and the position of the ROS construction site inspection robot device in the environment, combines the global path planner and the local path planner, generates a feedback signal to drive the movement of the vehicle body, and returns to step s22 until the ROS construction site inspection robot device reaches the target position.

[0208] The global path planner includes A ＊ algorithm.

[0209] The local path planner includes the DWA algorithm.

[0210] The global path planner uses A ★ algorithm, which is a commonly used search method for path finding and graph traversal to solve the shortest path. It combines the advantages of the Dijkstra algorithm and the best-first search algorithm (BFS). While improving the algorithm efficiency, it can ensure that the search is carried out along the path with the lowest expected cost. The local path planner uses the DWA algorithm, and the DWA algorithm mainly optimizes the speed of the robot during navigation so that the robot will neither collide with obstacles nor reach the target point in the shortest possible time.

[0211] Example 19:

[0212] A ROS construction site inspection robot device based on the integration of a multi-dimensional sensor array and an autonomous navigation optimization algorithm. The main technical content can be found in any one of Embodiments 11 to 18. Further, the ROS construction site inspection robot device is also provided with an error prompt module and a recovery behavior module.

[0213] In the absence of dynamic obstacles, if the ROS construction site inspection robot device cannot reach the target position, the error prompt module sends an error signal to the user.

[0214] When the upper computer detects that the ROS construction site inspection robot device is stuck by an obstacle and cannot move forward, the recovery behavior module executes a recovery behavior.

[0215] Recovery strategy: Integrate recovery strategy plugins such as rotate_recovery, move_slow_and_clear, and clear_costmap_recovery on the robot navigation function package. The robot completes the target pose within the tolerance range specified by the user. In the absence of dynamic obstacles, the navigation node will eventually make the robot reach the target point within the tolerance range or send an error signal to the user when it cannot reach the target location. When the robot feels stuck, the navigation node selects to execute a recovery behavior.

[0216] Embodiment 20:

[0217] A ROS construction site inspection robot device based on the integration of a multi-dimensional sensor array and an autonomous navigation optimization algorithm. The main technical content can be found in any one of Embodiments 11 to 19. Further, the ROS construction site inspection robot device also includes a human-machine interaction interface.

[0218] The human-machine interaction interface is used to display information about the open-air construction site environment and the construction site environment map.

[0219] Embodiment 21:

[0220] See Figures 1 to 8 , a ROS construction site inspection robot device based on the integration of a multi-dimensional sensor array and an autonomous navigation optimization algorithm. The main technical content includes:

[0221] I. Overall mechanical structure design and combination of functional hardware.

[0222] 1. Design of the body structure based on the operating intensity in a complex environment and the assembly conditions of multiple sensors:

[0223] S1: Concept Design: Based on the requirements analysis, a preliminary design plan for the fuselage is proposed, including determining the mechanical vehicle as the device carrier, determining the basic shape as a wheeled mechanical vehicle, determining the preliminary dimensions according to the design conditions, and selecting high-strength, lightweight carbon fiber composite materials as the fuselage material according to the complexity of the construction site environment to meet the strength and weight requirements.

[0224] S2: Structural Layout Design: Determine the positions and layouts of the components inside the fuselage, including sensors, power supplies, control systems, drive systems, etc., to ensure smooth connection and communication between them.

[0225] S3: Detail Design: Combining the layout design, elaborate on every detail of the fuselage, including connectors, fasteners, seals, etc., to ensure that they can stably and reliably fix various components of the fuselage in a complex environment.

[0226] S4: Preliminary Design Adjustment: Use software such as ADAMS and ANSYS to conduct a force analysis on the fuselage, determine the main stress-bearing parts and key nodes, and optimize the shape and dimensions to improve the strength and stiffness of the overall structure.

[0227] 2. Design of the Chassis Drive System Based on the Operating Speed and Requirements for Multiple Motion Postures in a Complex Environment:

[0228] Power Source Selection: According to the operating speed and load requirements, select a high-performance two-way motor drive and attach a compatible 12V polymer lithium battery energy storage battery.

[0229] Motion Mechanism Selection: Define the operating requirements of the chassis in a complex environment, including key parameters such as maximum speed, acceleration, obstacle-crossing ability, climbing angle, minimum turning radius, etc., and select a Mecanum omnidirectional wheel mechanism with a pendulum suspension that can achieve multiple motion postures.

[0230] 3. Selection of Multidimensional Sensors for Safety Detection Purposes.

[0231] Range Sensor: Adopt the SlAMTEC RPLIDAR A1 radar. Using the laser triangulation ranging technology it employs and cooperating with its independently developed high-speed vision acquisition and processing mechanism, it can perform more than 8,000 ranging operations per second to ensure fast and accurate environmental perception. Its compatible ROS system is convenient for integration with ROS robot systems.

[0232] Inertial Sensor: Adopt a 6-axis IMU sensor, which consists of three accelerometers and three gyroscopes. Each accelerometer and gyroscope is responsible for measuring the acceleration and angular velocity of an object in one axis in three-dimensional space respectively.

[0233] Vision sensor: Use a ROS depth camera to obtain the distance information of each point in the image from the camera. Combine the two-dimensional image coordinates to construct a three-dimensional space model to achieve scene modeling and object recognition. It can be easily integrated into the ROS system and use the algorithms and tools of ROS for data processing and decision-making.

[0234] Temperature and humidity sensor: Use a DHT11 digital temperature and humidity sensor developed based on the 51 development board and equipped with Bluetooth connection function.

[0235] 4. Selection of open-source controller and function development board.

[0236] Open-source controller: Use OpenCTR. Utilize its powerful control capabilities and scalability to support various chassis types and motor control, as well as software algorithm development through functions such as serial port or CAN connection to the outside world.

[0237] Function development board: Use the Raspberry Pi 4B development board. Utilize its high-performance hardware configuration and rich interface options for sensor connection and integration.

[0238] II. Development of algorithms such as sensor fusion and autonomous navigation based on the ROS system.

[0239] 1. System platform construction

[0240] Create a development environment for the ros robot in the virtual machine and create a file package in the directory to store ros function packages.

[0241] 2. Mapping algorithm

[0242] S1: Install the corresponding function packages in ROS (this invention includes algorithms such as gmapping, cartographer, hector_slam, etc.), and subscribe to the scan information of the lidar ( / scan topic) and the motion information of the robot ( / tf message for coordinate transformation) in the robot node.

[0243] S2: Data acquisition: Obtain environmental information through sensors such as lidar, including information such as the position and shape of obstacles in the environment.

[0244] S3: Data preprocessing: Preprocess the obtained environmental information, such as removing noise and smoothing, to improve the accuracy and reliability of the data.

[0245] S4: Particle filter: Based on the Monte Carlo method, approximately represent the pose distribution of the robot through a series of random samples (i.e., particles), and then estimate the position of the robot in the environment.

[0246] S5: Map construction: During the process of map construction, factors such as map resolution and scale are considered, and an accurate and easy-to-understand map is generated based on the pose estimation of the robot and sensor data.

[0247] 3. Autonomous Navigation Algorithm

[0248] S1: Install the corresponding functional package ORB_SLAM2 in ROS, subscribe to topics such as images, cameras, and sensors, and publish topics such as positioning information and map data key frames.

[0249] S2: Map construction: ORB-SLAM first uses the ORB feature extraction algorithm to extract key points and their descriptors from images. By matching feature points between two or more frames of images, the fundamental matrix or homography matrix is calculated, thereby estimating the initial position and pose of the camera and constructing an initial 3D point cloud map. As the robot continues to move and new images are acquired.

[0250] S3: Real-time positioning: ORB-SLAM attempts to track the feature points in the previous frame of image and estimates the current position and pose of the camera based on the tracking results. To balance computational efficiency and map quality, ORB-SLAM selectively retains some representative image frames as key frames for subsequent map construction and positioning.

[0251] S4: Path planning: Use a navigation node such as move_base to receive the map and positioning information provided by ORB-SLAM, and combine a global path planner and a local path planner to calculate the optimal path from the current position to the target position. Among them, the global path planner uses the A ★ algorithm, which is a commonly used search method for path finding and graph traversal to solve the shortest path. It combines the advantages of the Dijkstra algorithm and the best-first search algorithm (BFS), and while improving the algorithm efficiency, it can ensure that the search is carried out along the path with the lowest expected cost. The local path planner uses the DWA algorithm, and the DWA algorithm mainly optimizes the speed of the robot during navigation so that the robot neither collides with obstacles nor can reach the target point in the shortest possible time.

[0252] S5: Obstacle avoidance: The Cosmap cost map is used to achieve the perception and measurement of obstacles. Cosmap uses multiple independent rasterized layers to independently maintain obstacle information, and these layers can be superimposed to achieve a specific obstacle description layer. Each layer has its own obstacle update rules, such as adding obstacles, deleting obstacles, and updating the confidence of obstacle points, etc.

[0253] S6: Recovery Strategy: Integrate recovery strategy plugins such as rotate_recovery, move_slow_and_clear, and clear_costmap_recovery in the robot navigation function package. The robot completes the target pose within the tolerance range specified by the user. In the absence of dynamic obstacles, the navigation node will eventually make the robot reach the target point within the tolerance range or send an error signal to the user when the target location cannot be reached. When the robot senses that it is stuck, the navigation node selects to perform a recovery behavior.

[0254] III. Data Feedback

[0255] 1. Image Sharing: After using the Gmapping algorithm to create a map of the environment, rely on the map_server function package to store the created map. After the camera node is started, a web-based image server can be started, and the robot camera data can be displayed in the browser by entering the website address in the browser. First, connect to the robot through the SSH command, run the relevant camera node, and select the startup file. After successful startup, run the ROS command to start the web video server. Enter the relevant website address and select the corresponding topic to view the current image of the robot.

[0256] 2. Temperature and Humidity Information Feedback

[0257] S1: Data Acquisition: The single-chip microcomputer sends a startup signal by controlling the DATA pin of the DHT11. The DHT11 responds with a signal and sends 40-bit data to the single-chip microcomputer. The single-chip microcomputer receives the data and parses out the temperature and humidity values.

[0258] S2: Data Packing: The single-chip microcomputer converts the parsed temperature and humidity values into string format and sends the string data to the Bluetooth module through the serial port.

[0259] S3: Wireless Transmission: After receiving the serial port data, the Bluetooth module converts it into a Bluetooth signal and sends it out. The mobile phone App searches for and connects to the Bluetooth module through the Bluetooth function, receives the Bluetooth signal sent by the Bluetooth module, and converts it into a recognizable data format.

[0260] S4: Data Display: The mobile phone App parses the received data and displays it on the interface in real time. Users can intuitively see the temperature and humidity information of the current environment through the App interface.

Claims

1. A ROS site inspection robot device based on multi-dimensional sensor array integration and autonomous navigation optimization algorithm, characterized in that: It includes a mechanical vehicle body, a drive module, a multi-dimensional sensor detection module, a signal transmission module, and a host computer; The mechanical vehicle body is equipped with a driving module, a multi-dimensional sensor detection module, and a signal transmission module; The driving module drives the mechanical vehicle body to move according to the signal fed back by the host computer. The multi-dimensional sensor detection module is used to obtain information about the open-air construction site environment. The information of the open-air construction site environment at least includes an environment image; The signal transmission module is used to transmit the information of the open-air construction site environment acquired by the multi-dimensional sensor detection module to the host computer; The host computer constructs a construction site environment map based on the information of the open-air construction site environment, and plans the moving path of the ROS construction site inspection robot device based on the autonomous navigation optimization algorithm to generate a feedback signal; The host computer transmits the feedback signal to the drive module.

2. A ROS site inspection robot device based on multi-dimensional sensor array integration and autonomous navigation optimization algorithm according to claim 1, characterized in that: The mechanical vehicle body comprises a wheeled mechanical vehicle.

3. The ROS site inspection robot device based on multi-dimensional sensor array integration and autonomous navigation optimization algorithm according to claim 1 is characterized in that: The material of the mechanical vehicle body includes carbon fiber composite material.

4. The ROS site inspection robot device based on multi-dimensional sensor array integration and autonomous navigation optimization algorithm according to claim 1 is characterized in that: The driving module includes a power source and a motion mechanism; The power source provides power for the motion mechanism; The power source includes two motor drive modules and an energy storage battery; The energy storage battery provides energy for the two-way motor drive modules; One motor drive module is used to drive the movement of the mechanical vehicle body, and the other motor drive module is used to drive the motion mechanism; The motion mechanism is used to drive the mechanical vehicle body to rotate; The motion mechanism includes a Mecanum omnidirectional wheel mechanism.

5. The ROS site inspection robot device based on multi-dimensional sensor array integration and autonomous navigation optimization algorithm according to claim 1 is characterized in that: The multi-dimensional sensor detection module includes a distance measuring sensor, an inertial sensor, a visual sensor, and a temperature and humidity sensor; The distance measuring sensor is used to detect the distance between the obstacle and the ROS site inspection robot device and the distance between the ROS site inspection robot device and the target position; The target position is the end point of the moving path of the ROS site inspection robot device; The inertial sensor is used to monitor the motion state of the inspection robot device in real time and provide position information; The visual sensor is used to obtain images of the surrounding environment of the ROS construction site inspection robot device; The temperature and humidity sensor is used to detect ambient temperature and humidity.

6. The ROS site inspection robot device based on multi-dimensional sensor array integration and autonomous navigation optimization algorithm according to claim 1 is characterized in that: The signal transmission mode of the signal transmission module includes Bluetooth transmission and serial port transmission.

7. The ROS site inspection robot device based on multi-dimensional sensor array integration and autonomous navigation optimization algorithm according to claim 1 is characterized in that: The steps of constructing a construction site environment map based on the information of the open-air construction site environment by the host computer are as follows: S11 obtains information about the open-air construction site environment through a multi-dimensional sensor detection module; s12 preprocesses the acquired information of the open-air construction site environment to obtain preprocessed data; The preprocessing includes noise removal, smoothing, and feature extraction; s13 performs particle filtering on the preprocessed data based on the Monte Carlo method to evaluate the position of the ROS site inspection robot device in the environment; s14 constructs a construction site environment map based on the assessed location and information about the open-air construction site environment.

8. The ROS site inspection robot device based on multi-dimensional sensor array integration and autonomous navigation optimization algorithm according to claim 7 is characterized in that: The host computer plans the moving path of the ROS site inspection robot device based on the autonomous navigation optimization algorithm, and the steps of generating the feedback signal are as follows: s21 sets the initial position and target position of the ROS site inspection robot device; The target position is the end point of the moving path of the ROS site inspection robot device; S22 builds a real-time construction site environment map based on the information of the open-air construction site environment obtained by the multi-dimensional sensor detection module, and evaluates the position of the ROS construction site inspection robot device in the environment; s23 determines whether there are obstacles in the surrounding environment of the ROS site inspection robot device. If yes, the Cosmap cost map is used to avoid the obstacles. If not, it goes to step s24; The steps of using the Cosmap cost map to avoid obstacles are as follows: s231 obtains the conversion relationship between the global coordinate system and the robot coordinate system, and loads the layer; The layers include a static map layer, an obstacle map layer, and an expansion layer; s232 updates the obstacle map layer in real time through the information of the open-air construction site environment and the motion status of the ROS construction site inspection robot device, and calculates the expansion layer based on the static map layer and the obstacle map layer; s233 transmits the updated data to step s24; s24 calculates the optimal path from the current position to the target position based on the real-time construction site environment map and the position of the ROS construction site inspection robot device in the environment, combines the global path planner and the local path planner, and generates a feedback signal to drive the mechanical vehicle body to move, and returns to step s22 until the ROS construction site inspection robot device reaches the target position; The global path planner includes A ＊ algorithm; The local path planner includes a DWA algorithm.

9. The ROS site inspection robot device based on multi-dimensional sensor array integration and autonomous navigation optimization algorithm according to claim 8 is characterized in that: The ROS site inspection robot device is also provided with an error prompt module and a recovery behavior module; In the absence of dynamic obstacles, if the ROS site inspection robot device cannot reach the target position, the error prompt module sends an error signal to the user; When the host computer detects that the ROS site inspection robot device is stuck by an obstacle and cannot move forward, the recovery behavior module executes the recovery behavior.

10. The ROS site inspection robot device based on multi-dimensional sensor array integration and autonomous navigation optimization algorithm according to claim 1 is characterized in that: The ROS site inspection robot device also includes a human-machine interaction interface; The human-computer interaction interface is used to display information about the open-air construction site environment and a construction site environment map.

Images