Complex underground space terrain exploration vehicle

Abstract

The invention provides a complex underground space terrain exploration vehicle which comprises a drive-by-wire chassis, a navigation module, a laser radar, a visual image recognition module, a wireless communication module and a master control system, the drive-by-wire chassis adopts a four-wheel frame, and the navigation module, the laser radar, the visual image recognition module, the wireless communication module and the master control system are installed on the four-wheel frame. The master control system is an intelligent driving system, and the intelligent driving system obtains the satellite position and attitude measurement of the vehicle by using navigation module data; meanwhile, the intelligent driving system fuses data of a navigation module, a laser radar and a visual image recognition module to carry out SLAM intelligent mapping; through an obstacle avoidance algorithm and a vehicle driving decision algorithm, vehicle obstacle avoidance, vehicle path planning and driving decision making are realized. The invention provides a complex underground space terrain exploration vehicle, which realizes autonomous obstacle avoidance, and is higher in routing inspection planning capability, higher in accuracy and higher in recognition capability.

Description

Technical Field

[0001] The invention belongs to the field of electronic information technology, and in particular relates to a complex underground space terrain exploration vehicle. Background Art

[0002] Intelligent inspection robots are based on intelligent technology and are programmable. With the development of technology, the functions of intelligent inspection robots have been gradually improved, allowing them to simulate manual operations during application and replace traditional manual inspections. This inspection method is more flexible and less affected by external climate, environment and other factors, and is suitable for various inspections. With the development of science and technology, inspection robots will have higher intelligence and autonomy, and better adapt to various environments and tasks.

[0003] Existing technology: Intelligent inspection robots have low adaptability to complex environments. Although they have certain intelligence and automation capabilities, their adaptability to complex environments is relatively low. Current robot technology cannot fully adapt to these environments, especially the support capabilities in underground spaces such as light, humidity, low oxygen concentration, and large-span underground spaces.

[0004] At the same time, in complex underground spaces, there is no light, no satellites, weak radiation, and the environmental space conditions and size are unknown. Traditional inspection robots used in underground spaces have their related sensors and hardware equipment fixed on them. When encountering narrow or low areas, the sensors installed on the inspection robots are easily bumped and damaged. Summary of the invention

[0005] In response to the problems existing in the application of current intelligent inspection robots in underground spaces, the present invention provides a complex underground space terrain exploration vehicle with real-time positioning and map building capabilities, stronger autonomous obstacle avoidance and planning, higher accuracy, and stronger recognition capabilities.

[0006] The solution adopted by the present invention to solve its technical problems is: a complex underground space terrain exploration vehicle, including a wire-controlled chassis, a navigation module, a laser radar, a visual image recognition module, a wireless communication module and a main control system. The wire-controlled chassis adopts a four-wheel frame, and the navigation module, the laser radar, the visual image recognition module, the wireless communication module and the main control system are installed on the four-wheel frame. The main control system is an intelligent driving system, and the intelligent driving system uses the navigation module data to obtain the vehicle's satellite position and attitude; at the same time, the intelligent driving system integrates the navigation module, the laser radar and the visual image recognition module data to perform SLAM intelligent mapping; through the obstacle avoidance algorithm and the vehicle driving decision algorithm, the vehicle's obstacle avoidance, vehicle path planning and driving decision are realized.

[0007] The four-wheel frame is equipped with a patrol mechanism, which includes a base plate, a motor, a synchronous wheel, a synchronous belt, a lead screw, a slider, a fixed shaft, a pull rod and a flipping mechanism. Parallel lead screws are fixed on both sides of the base plate, a slider is mounted on the lead screw, a synchronous wheel is installed at the end of the lead screw, a motor is installed between the parallel lead screws, a driving wheel is installed on the motor output shaft, and the driving wheel is connected to the synchronous wheels on both sides through a synchronous belt drive, thereby driving the lead screw to rotate synchronously and driving the slider to move synchronously. A flipping mechanism is installed on the slider, and a laser radar and a visual image The identification module is installed on the flipping mechanism, which includes a front rod, a rear rod, a lower connecting rod and an upper platform. A group of shaft seats parallel to the lead screw are arranged on the slider, and the front rod and the rear rod are hinged on the shaft seats respectively. The upper ends of the front rod and the rear rod are installed with the upper platform, and the lower connecting rod is installed on the lower side of the front rod and the rear rod. The front rod and the rear rod are parallel, and the upper platform and the lower connecting rod are parallel. A fixed shaft is arranged at the end of the bottom plate, and pull rods are installed at both ends of the fixed shaft. The other ends of the pull rods are hinged to the corresponding front rods respectively, and the length of the pull rods is greater than the distance from the fixed shaft to the lead screw.

[0008] Furthermore, a visual image recognition module is installed on the upper platform, and a laser radar is installed between the front poles on both sides. The laser radar uses a multi-line laser radar sensor, and uses the high-precision registration of the point cloud data collected by it and the image collected by the visual sensor to achieve effective processing and understanding of the laser point cloud data, thereby improving the complex environment perception and analysis capabilities of the equipment.

[0009] Furthermore, the intelligent driving system uses bundle adjustment to optimize the camera pose and landmark points. The intelligent driving system obtains the camera intrinsic parameter matrix, extrinsic parameter matrix, image feature points and initial values ​​of spatial 3D object point coordinates in the visual image recognition module, and uses the BA model to optimize the above parameters as a whole, eliminate errors, and obtain accurate visual information.

[0010] Furthermore, the data collected by the intelligent driving system is fused and processed using an extended Kalman filter. The filter predicts the state based on the wheel speed odometer and navigation module data. The navigation module uses IMU-UWB fusion navigation and visual navigation to obtain the estimated value of the vehicle's position and attitude.

[0011] Beneficial effects of the present invention: The present invention provides a complex underground space terrain exploration vehicle. The intelligent patrol robot provides instant positioning and map building capabilities through laser radar, navigation module and visual image recognition module, and realizes stronger autonomous obstacle avoidance and patrol planning capabilities, higher accuracy, and stronger recognition capabilities. It can serve special scenarios and is suitable for underground space environments. It solves the problems that similar patrol robots cannot work in low-light environments, large scene changes, large-span underground spaces and other environments, and at the same time optimizes and improves the operating capabilities in ionizing radiation environments.

[0012] At the same time, the present invention further provides a flipping mechanism, through which the intelligent patrol robot can provide height adjustment and protection for the laser radar, visual image recognition module, etc. in a narrow space, avoiding hardware damage due to bumps in unknown narrow spaces, and improving the reliability of the patrol robot in use. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 It is a schematic diagram of the overall structure of the present invention.

[0014] Figure 2 It is a three-dimensional structural schematic diagram of the flip mechanism.

[0015] Figure 3 It is a side sectional structural schematic diagram of the flip mechanism.

[0016] Figure 4 It is the frame view of the intelligent driving system.

[0017] Figure 5 It is a schematic diagram of the intelligent driving system design.

[0018] Numbers in the figure: four-wheel frame 1, patrol mechanism 2, visual image recognition module 3, laser radar 4, base plate 21, motor 22, synchronous wheel 23, synchronous belt 24, screw 25, slider 26, fixed shaft 27, pull rod 28, flip mechanism 29, front rod 291, rear rod 292, lower connecting rod 293, upper platform 294. DETAILED DESCRIPTION

[0019] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be described in detail below.

[0020] Embodiment 1: The present invention provides a complex underground space terrain exploration vehicle device, which provides instant positioning and map building capabilities, autonomous obstacle avoidance and planning, higher accuracy, and a new type of intelligent patrol robot with stronger recognition capabilities. It serves special scenarios and is suitable for underground space environments. It solves the problems that similar patrol robots cannot work in low-light environments, large scene changes, large-span underground spaces, etc., and at the same time optimizes and improves its operating capabilities in ionizing radiation environments.

[0021] Specifically, the complex underground space terrain exploration vehicle includes a wire-controlled chassis, a navigation module, a laser radar 4, a visual image recognition module 3, a wireless communication module and a main control system, such as Figure 1 As shown, the wire-controlled chassis adopts a four-wheel frame 1, on which a navigation module, a laser radar 4, a visual image recognition module 3, a wireless communication module and a main control system are installed. The main control system is an intelligent driving system, and the intelligent driving system uses the navigation module data to obtain the vehicle's satellite position and attitude.

[0022] The four-wheel frame 1 is provided with a patrol mechanism 2. Figure 2 and Figure 3 As shown, the patrol mechanism 2 includes a base plate 21, a motor 22, a synchronous wheel 23, a synchronous belt 24, a lead screw 25, a slider 26, a fixed shaft 27, a pull rod 28 and a flipping mechanism 29. Parallel lead screws 25 are fixed on both sides of the base plate 21, and a slider 26 is mounted on the lead screw 25. The end of the lead screw 25 is equipped with a synchronous wheel 23. The motor 22 is installed between the parallel lead screws 25. The drive wheel is installed on the output shaft of the motor 22. The drive wheel drives the synchronous wheels 23 on both sides through the synchronous belt 24, thereby driving the lead screw to rotate synchronously and driving the slider 26 to move synchronously. The flipping mechanism 29 is installed on the slider 26, and the laser radar 4 and the visual image recognition module 3 are installed on the flipping mechanism 29. The mechanism 29 includes a front rod 291, a rear rod 292, a lower connecting rod 293 and an upper platform 294. A group of shaft seats parallel to the lead screw 25 are arranged on the slider 26, and the front rod 291 and the rear rod 292 are hinged on the shaft seats respectively. The upper ends of the front rod 291 and the rear rod 292 are installed with the upper platform 294, and the lower side of the front rod 291 and the rear rod 292 is installed with the lower connecting rod 293. The front rod 291 and the rear rod 292 are parallel, and the upper platform 294 and the lower connecting rod 293 are parallel. A fixed shaft 27 is arranged at the end of the base plate 21, and pull rods 28 are installed at both ends of the fixed shaft 27. The other ends of the pull rods 28 are hinged on the corresponding front rod 291 respectively, and the length of the pull rod 28 is greater than the distance from the fixed shaft 27 to the lead screw 25.

[0023] The flip mechanism 29 forms a parallelogram structure on the slider 26 through the front rod 291, the rear rod 292, the lower connecting rod 293 and the upper platform 294, so as to ensure that the upper platform 294 at its upper end can always maintain a horizontal state. When the motor 22 is working, the output shaft of the motor 22 will drive the driving wheel installed thereon, and the driving wheel transmits power to the synchronous wheels 23 on both sides through the synchronous belt 24, so that the parallel lead screws 25 rotate synchronously. The rotation of the lead screw 25 drives the slider 26 mounted thereon to move synchronously, thereby realizing the moving function of the device. At this time, the pull rod 28 on the bottom plate 21 limits the position of the front rod 291, pulls the flip mechanism 29 to tilt, thereby changing the height of the upper platform 294 of the flip structure, adapting to the passage of the narrow underground space, and avoiding the laser radar 4 and the visual image recognition module 3 installed on the upper platform 294 from being bumped.

[0024] The hardware system of the intelligent inspection robot of the present invention is mainly composed of a wire-controlled chassis, a navigation module, a laser radar 4, a visual image recognition module 3, a wireless communication module and a main control system. The software system is mainly composed of an intelligent driving system. The intelligent driving system is responsible for satellite positioning, vehicle attitude measurement, SLAM mapping, intelligent obstacle avoidance, image recognition and intelligent driving decision-making functions.

[0025] The wire-controlled chassis adopts a four-wheel frame 1, which is the hardware basis of the entire system. Other functional components are attached and installed on the four-wheel frame 1. The four-wheel frame 1 is mainly composed of a chassis frame, a drive system, a steering system, and a bus system. While completing the control commands sent by the main control, it also provides DC power supply to other hardware.

[0026] The navigation module performs satellite positioning of the vehicle's position and real-time measurement of the vehicle's current posture, and transmits the measurement to the main control system after packaging through a custom protocol for use by other functional nodes. The laser radar 4 uses laser time-of-flight ranging technology to calculate the distance between the obstacle and the vehicle by measuring the time difference between laser emission and reflection, and combines the internal DSP processor for real-time solution to achieve the purpose of ranging. After transmitting the laser scanning point cloud data to the main control system, it is processed by the obstacle avoidance algorithm to achieve obstacle avoidance. The visual image recognition module 3 collects the depth information of the object in front and realizes image recognition after processing by the image recognition algorithm. The wireless communication module transmits the packaged data protocol to the corresponding port of the server through the TCP protocol for use by the location service system. The main control system is responsible for running the intelligent driving system, and uniformly converts the time and coordinates of the data collected by each sensor device, and transmits the control command to the wire-controlled chassis through the CAN bus.

[0027] The intelligent driving system processes the data collected by each hardware system and packages it in a defined data format for use by each functional module. The vehicle's satellite position and attitude are measured using the data from the inertial satellite combined navigation module; SLAM intelligent mapping is achieved using the LiDAR 4 and inertial satellite navigation module; and the vehicle's obstacle avoidance, vehicle path planning and driving decision-making are achieved through intelligent obstacle avoidance algorithms and vehicle driving decision-making algorithms.

[0028] Specific usage: Hardware testing: for the equipment used, including appearance, interface, structure, circuit selection and function, the hardware test is divided into performance sequencing, appearance inspection, interface test, power supply test, and functional verification; the purpose is to test whether the hardware equipment has good performance, avoid problems in the later process, and solve problems in time; improve product quality. This project mainly uses Shinto, JMeter tools, and Loadrunner performance automation tools for hardware testing; the test equipment includes oscilloscopes, voltage regulators, multimeters, digital voltages, and calculators. Hardware testing mainly verifies the technical indicators and functionality of the main functional modules of the product to ensure high-quality and high-quality completion of the hardware equipment. The main hardware equipment of this product includes linear chassis, industrial control systems, obstacle avoidance radars, and combined navigation.

[0029] Software testing: The software and hardware are systematically coordinated to ensure the overall high-quality and high-performance operation to achieve the desired purpose; the operation of the equipment is controlled by software, and the interface performance is tested through single touch and multiple clicks of buttons, and the collected data is collected and stored, and the displayed effects are rationalized and corrected; through continuous regression testing, correction, improvement, and complete software operating instructions and steps, the desired effect is achieved.

[0030] System testing: This operating system can use hardware equipment and build an environment for verification and operation; and ensure that it can run efficiently, flexibly, and with clear instructions. The installation steps, operation manuals, experimental manuals, and design requirements will be displayed through system testing to achieve the results. When there are missing items of information, the system will be integrated and the required test data will be provided; when the hardware platform is not available, demonstration data will be provided; when some public data is missing in the test system, some data for testing or demonstration will be provided; problems encountered during the test will be recorded, analyzed, and problems will be found and corrected; the design requirements will be used as a reference standard to implement the corresponding functions.

[0031] Compliance testing: This is conducted by a dedicated testing team, which consists of experienced system designers and programmers. The entire testing activity is conducted in the presence of reviewers. After completing the scheduled assembly test work, the testing team should be responsible for collating and analyzing the test results and forming a test report. The test report records the actual test results, problems found during the test, methods for solving these problems, and the results of retesting after solving them.

[0032] Embodiment 2: The intelligent inspection robot provided by the present invention is mainly used to solve and meet the needs of finely constructing underground space three-dimensional maps and target object locations and tracking in large spans and complex underground environments. The device integrates a multi-line laser radar 4 sensor, and uses the high-precision registration of the point cloud data collected by it and the images collected by the visual sensor, that is, accurately determining the pixel coordinates of the three-dimensional point cloud on the digital image, so as to achieve effective processing and understanding of the laser point cloud data and enhance the complex environment perception and analysis capabilities of the device. Its main application technology is the beam adjustment registration technology, which is a new technical means and method built on the visual SLAM technology.

[0033] The bundle adjustment (BA) technology is mainly used for the back-end optimization of camera pose and landmarks. After obtaining the camera intrinsic matrix, extrinsic matrix, image feature points and initial values ​​of spatial 3D object point coordinates, the BA model can be used to optimize the above parameters as a whole, eliminate errors, and obtain accurate visual information. In visual reconstruction, the goal of the BA model is to obtain more accurate camera pose and spatial landmarks. All sensor data are fused and processed using an extended Kalman filter. The filter predicts the state based on the wheel speed odometer and IMU data. UWB and visual navigation are used to obtain the position and attitude estimation of the vehicle body. The fusion positioning and attitude determination method composed of dual-label UWB, IMU, LiDAR 4 and other sensors, IMU can obtain continuous high-frequency system prediction status, can judge UWB non-line-of-sight conditions, dual-label UWB can solve the vehicle heading, so that the system has complete position and attitude measurement information, and cooperates with high-precision multi-line LiDAR 4 and UWB indoor positioning to complete indoor positioning and complex underground space scene three-dimensional map construction and intelligent navigation.

[0034] Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other implementation methods obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

Claims

1. A complex underground space terrain exploration vehicle, comprising a wire-controlled chassis, a navigation module, a laser radar (4), a visual image recognition module (3), a wireless communication module and a main control system, characterized in that: The control-by-wire chassis adopts a four-wheel frame (1), on which a navigation module, a laser radar (4), a visual image recognition module (3), a wireless communication module and a main control system are installed. The main control system is an intelligent driving system. The intelligent driving system uses navigation module data to obtain the satellite position and attitude of the vehicle. At the same time, the intelligent driving system integrates the navigation module, the laser radar (4) and the visual image recognition module (3) data to perform SLAM intelligent mapping. By means of an obstacle avoidance algorithm and a vehicle driving decision algorithm, the vehicle's obstacle avoidance, vehicle path planning and driving decision are achieved. The four-wheel vehicle frame (1) is provided with a patrol mechanism (2), the patrol mechanism (2) comprising a base plate (21), a motor (22), a synchronous wheel (23), a synchronous belt (24), a lead screw (25), a slider (26), a fixed shaft (27), a pull rod (28) and a flip mechanism (29), parallel lead screws (25) are fixed on both sides of the base plate (21), the slider (26) is mounted on the lead screw (25), the synchronous wheel (23) is mounted on the end of the lead screw (25), the motor (22) is mounted between the parallel lead screws (25), the drive wheel is mounted on the output shaft of the motor (22), the drive wheel drives the synchronous wheels (23) connected on both sides through the synchronous belt (24), thereby driving the lead screw to rotate synchronously, driving the slider (26) to move synchronously, the flip mechanism (29) is mounted on the slider (26), the laser radar (4) and the visual image recognition module (3) are mounted on the flip mechanism (2 9), the flip mechanism (29) comprises a front rod (291), a rear rod (292), a lower connecting rod (293) and an upper platform (294), a group of shaft seats parallel to the lead screw (25) are arranged on the slider (26), the front rod (291) and the rear rod (292) are respectively hinged on the shaft seats, the upper ends of the front rod (291) and the rear rod (292) are installed with the upper platform (294), the front rod (291) and the rear rod (292) A lower connecting rod (293) is installed on the lower side, the front rod (291) and the rear rod (292) are parallel, the upper platform (294) and the lower connecting rod (293) are parallel, a fixed shaft (27) is provided at the end of the bottom plate (21), and pull rods (28) are installed at both ends of the fixed shaft (27), and the other ends of the pull rods (28) are respectively hinged on the corresponding front rod (291), and the length of the pull rod (28) is greater than the distance from the fixed shaft (27) to the lead screw (25).

2. The complex underground space terrain exploration vehicle according to claim 1, characterized in that: A visual image recognition module (3) is installed on the upper platform (294), and a laser radar (4) is installed between the front bars (291) on both sides.

3. The complex underground space terrain exploration vehicle according to claim 2, characterized in that: The laser radar (4) uses a multi-line laser radar (4) sensor, and utilizes the high-precision registration of the point cloud data collected by the multi-line laser radar (4) and the images collected by the visual sensor to achieve effective processing and understanding of the laser point cloud data, thereby improving the complex environment perception and analysis capabilities of the device.

4. The complex underground space terrain exploration vehicle according to claim 1, characterized in that: The intelligent driving system uses bundle adjustment to optimize the camera pose and landmark points. The intelligent driving system obtains the camera intrinsic parameter matrix, extrinsic parameter matrix, image feature points and initial values ​​of spatial 3D object point coordinates in the visual image recognition module (3), and uses the BA model to optimize the above parameters as a whole, eliminate errors, and obtain accurate visual information.

5. The complex underground space terrain exploration vehicle according to claim 1, characterized in that: The data collected by the intelligent driving system is fused and processed using an extended Kalman filter. The filter predicts the state based on the wheel speed odometer and navigation module data. The navigation module uses IMU-UWB fusion navigation and visual navigation to obtain the estimated value of the vehicle's position and attitude.

Images