CN113465728B - Terrain awareness method, system, storage medium and computer equipment - Google Patents
Terrain awareness method, system, storage medium and computer equipment Download PDFInfo
- Publication number
- CN113465728B CN113465728B CN202110710851.5A CN202110710851A CN113465728B CN 113465728 B CN113465728 B CN 113465728B CN 202110710851 A CN202110710851 A CN 202110710851A CN 113465728 B CN113465728 B CN 113465728B
- Authority
- CN
- China
- Prior art keywords
- robot
- algorithm
- laser
- path
- terrain
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01H—MEASUREMENT OF MECHANICAL VIBRATIONS OR ULTRASONIC, SONIC OR INFRASONIC WAVES
- G01H17/00—Measuring mechanical vibrations or ultrasonic, sonic or infrasonic waves, not provided for in the preceding groups
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
Abstract
The invention belongs to the technical field of terrain perception, and discloses a terrain perception method, a system, a storage medium and computer equipment, wherein data information about terrain parameters is obtained by using a Kinect depth camera, so that automatic recognition of road surface danger level is realized; removing the motion distortion of the laser radar by combining an odometer auxiliary method with a Kalman filtering algorithm; optimization of SLAM mapping and navigation algorithms in ROS. In the invention, the distance error of the robot global positioning on the X axis and the Y axis is less than 10cm, and the angle error is less than 0.2rad; the dimension error of the grid map and the real environment is smaller than 1cm; the deviation degree of the real navigation path of the robot and the algorithm estimated path is smaller than 10cm; the mobile robot can perform loop-back when building a graph in an environment with similar structure; the variance of the point cloud data acquired by the laser radar after pretreatment is smaller than 0.1.
Description
Technical Field
The invention belongs to the technical field of terrain awareness, and particularly relates to a terrain awareness method, a system, a storage medium and computer equipment.
Background
At present, the coal resources in China are rich, and the coal is still a main energy source in China. In recent years, the coal industry sequentially brings forward ideas such as mechanical man changing, intelligent unmanned and the like. In 1 month 2019, the national coal mine safety supervision bureau issues a focus development catalog of coal mine robots, and publishes 5 kinds of 38 kinds of coal mine robots which are used for focus development of coal enterprises, scientific research institutions and manufacturing enterprises, wherein the inspection robot is one of safety control robots.
In recent years, the national increasing force is used for renovating and standardizing the intrinsic safety type management of coal mines, and simultaneously, coal mine enterprises are greatly promoted to develop in the direction of digitalization and unmanned. The inspection work of equipment in any industry is generally finished by manpower, a large amount of human resources are consumed, the work flow is complex, particularly in high-risk industries, risks can exist in the inspection process, and the working environment of personnel is poor. The underground main drainage system of the coal mine is an indispensable important device in coal production, and the operation condition of the underground main drainage system directly influences the normal production of the mine and the life safety of workers, so that in order to ensure the normal operation of a water pump, patrol staff need to check the operation condition of the device regularly, such as the conditions of a water pump packing, a water sump condition, a motor, whether the water pump is abnormal or not, and the like. In general, a water pump house needs several people to watch and operate, and has low working efficiency.
Along with the progress of scientific technology, in the electric power and petrochemical industry, wheeled inspection robots, crawler-type inspection robots or rail-type hanging inspection robots have many application cases, and mobile robots have strong pertinence inspection on inspected equipment in the inspection process, and the working quality is even more specialized than people.
The coal mine disasters are frequent, the hazard is serious, the risk is large, the underground workers are dense, the dangerous posts are numerous, the research and development application of the coal mine water pump house inspection robot can replace inspection workers to carry out equipment and environment inspection, the labor intensity of the workers can be effectively reduced, the potential safety hazard existing in the inspection process is reduced, the inspection quality is improved, and meanwhile, the intrinsic safety level of coal mine enterprises is improved to the greatest extent. The method has important significance for pushing the technological revolution of coal exploitation, realizing the high-quality development of coal industry, pushing the intelligent progress of coal mine and national energy safety supply.
From the application environment of robots, domestic and foreign specialists divide robots into two main categories, namely industrial robots in manufacturing environments and special robots in non-manufacturing environments. Research and development of special robots at home and abroad are mainly focused on the fields of exploration and relief, military explosion prevention, reconnaissance detection and the like. The development process of the special mobile robot for realizing the special functions in the mine is slower due to the special working environment of the mine and the high-standard mobile, communication and explosion-proof performance requirements.
(1) The coal mine detection type robot is designed by the Australian Federal department of Construction (CSIRO) in the 90 s of the last century, and the underground gas environment is the main work of the robot. The underground information is transmitted back to the ground through the visible light and infrared image sensor, and the latest underground state is informed to rescue team members so as to facilitate the establishment of rescue schemes. The power source of the robot is a 140Ah nickel-cadmium battery, the walking relies on a differential eight-wheel structure consisting of 750W motors assembled on two sides respectively, the size of the robot is 2.5 multiplied by 1.65m, the speed can reach two kilometers per hour at the highest speed, and the robot has the cruising ability of 8 hours. The Groundhog coal mine detection robot is developed by the university of Carniken university robot research institute, and is mainly used for underground environment detection and underground three-dimensional map drawing, and is provided with a laser sensor, a night vision camera, a somatosensory sensor, a gyroscope and the like. The four wheels are driven by hydraulic pressure and run in differential speed, and the steering system has the in-situ steering function. Thhun et al detected the main roadway of Ma Disi mine from pennsylvania cotini at 5/30 of 2003 and completed the three-dimensional pattern. In 2005, the first coal mine detection type robot CUMT-I of China is successfully developed by the university rescue and equipment research institute of China, can detect the environment of post-disaster sites, is matched with a dangerous gas sensor, a low-illuminance camera and a two-way voice intercom system, is controlled in a remote control mode, can randomly load various disaster relief materials (medicines, physical objects or rescue tools), and is a good search and rescue helper after disaster relief. In 2007, the university of Harbin industrial robot institute and the electric appliance company of Tangshan Kacheng developed a coal mine detection robot together, which can detect the environment of post-disaster site, and was matched with a dangerous gas sensor, a low illuminance camera and a two-way voice intercom system. The walking mode is crawler type, and the crawler is divided into three parts of driving, swinging arms and swinging legs. Remote control operation is supported, and remote sensing and control keys are used for realizing.
(2) The Ratler coal mine rescue robot is researched and developed by the United states department of labor Mine Safety and Health Administration (MSHA) and Mordi sub-Intelligent System and Robot Center (ISRC) together, is mainly used for post-disaster investigation, is provided with an infrared camera, a gyroscope and a dangerous gas sensor, and can be remotely controlled by a wireless radio frequency device for a maximum distance of 76 meters. The suitability experiment of the robot in the fire rescue of the Willow Greek coal mine in 12 th 1998 shows that the robot can not reach the standard of the coal mining industry on the rescue robot. The second generation mobile coal mine disaster relief robot platform Cave Crawler is designed by a university of Carniken Meilon robot research institute, a gear differential mechanism similar to a Mars detection vehicle with a 'courage number' is used in the robot, a rocker type left and right wheel moving system is connected with a robot main body through the differential mechanism, so that the swing angles of the two rockers are linearly averaged, and the swing angles of the robot main body are further converted into the swing angles of the robot main body, so that the balance of the robot main body is kept. When a certain side wheel is lifted, the swing amplitude of the whole vehicle is only half of the lifting amplitude of the wheel, so that the obstruction of the topography fluctuation to the movement of the robot is weakened. In addition, the design evenly distributes the weight of the vehicle body to each wheel of the robot, and the wheels can adjust the relative positions according to the terrain, so that the robot moves more stably, is not easy to turn over, and has stronger obstacle crossing capability.
(3) The coal mine inspection robot Zhou Zhan mainly researches the mechanical body of the underground roadway suspension line inspection robot, and further researches the control system of the inspection robot on the basis of carrying out detailed analysis on the body structure, autonomous movement and inspection process control requirements of the suspension line inspection robot. Wang Zhi is equally designed with a coal mine detection robot capable of autonomously extending communication distance and automatically retracting so as to break the limit of the underground coal mine detection robot on the receiving of working stroke. When the weakening of the wireless control signal is detected, the repeater ejection system carried by the robot can operate to prolong the wireless communication distance. If the wireless signal is terminated, the robot changes back to start a back-pulling program, and the robot returns to a safe position with smooth communication through the original path according to the sensor data recorded by operation. The robot realizes the retraction program by using the method of integrating the encoder and the laser sensor data, and the coordinate and the orientation of the robot are corrected by doubly matching the data according to the ICP algorithm, so that the sufficient retraction precision is achieved. From experimental results, the robot has the clear characteristics of longer working stroke and retractive capacity, which means that the robot can replace rescue workers to go deep into a mine to finish the initial work of hazard detection when the safety of a coal mine disaster site is unknown.
Jiang Junying and the like are designed to solve the problems of complex structure and control, low reliability, high cost and large occupied space of the traditional coal mine tunnel wheel type or crawler type robot. They focused on the mechanical design of this robot and built a model of a virtual prototype in Adams software for simulation analysis to assess its ability to cross obstacles in three states of motion, horizontal walking, up and down slopes. The result shows that the robot is relatively stable in all three motion states, particularly the robot always keeps uniform linear motion in the horizontal direction, and fluctuation within the allowable range can be generated in the vertical and horizontal side swinging directions. Zhou Mingjing for solving the inspection and monitoring problems of the underground belt conveyor, a track type unmanned inspection device is designed, and the overall design scheme, capability characteristics and software of the robot are described. Hardware development pathways. The inspection of large-scale equipment such as a coal mine underground belt conveyor, a scraper conveyor and the like can be completed through the inspection robot, so that unmanned on duty is truly and fundamentally realized, the burden of inspection personnel is reduced, and the safety management of a coal mine is promoted to the greatest extent.
Cui Xin on the basis of the study on the problems of workers in underground water pump rooms of coal mines, an automatic inspection robot device capable of replacing workers in inspection is designed, the design basis, the constitution and the functions of a robot system are described, the implementation of software and hardware of the robot, data acquisition and other modes are described, and the system can discover the possible problems of underground water pump rooms in real time and ensure the stable and safe operation of the system. Pei Wen and the like are used for designing an intelligent robot which has high reliability and can replace manual inspection, information acquisition, video inspection, obstacle reporting and other functions under the condition of carrying out inspection and supervision on the key of a water pump house in a coal mine to generate problem study, describing the constitution and the action characteristics of the robot equipment, and further describing the automatic inspection function of the robot under different conditions in the study conclusion, thereby opening a new mode for achieving unmanned duty and automatic management mode in the coal mine.
However, the prior art has the problems of map building and autonomous navigation; the mining water pump house inspection robot encounters severe natural conditions and complex geological environments, for example, various barriers such as accumulated water, cables, falling rocks and crushed coal are often arranged in a roadway; the roadway surface gradient is large, and the like, so that the inspection robot can establish a two-dimensional map of surrounding obstacles, and also can establish a two-dimensional map of rugged topography.
The robot can not encounter a slope road section in the walking process, and as the two-dimensional laser radar is used for building a map aiming at an obstacle on a horizontal plane in a space, the slope is regarded as an obstacle, so that the robot can not pass through the slope road section, and therefore, how to combine laser radar data (two-dimensional) and depth camera data (three-dimensional) to build a two-dimensional environment map of the slope by the inspection robot.
Autonomous navigation problem: the inspection robot is provided with a camera or an infrared thermal imager to observe the equipment, and the size of the electrical equipment in the observation field determines the quality of data acquisition precision, so that the inspection robot not only shows the characteristics of avoiding obstacles, having shortest walking path, having least time and the like when carrying out global path planning, but also needs to select proper observation distance and observation angle to determine the proper observation range of each electrical equipment.
Under the premise of ensuring the inspection safety, the inspection robot avoids static obstacles in the map from the inspection starting point and safely reaches the target point, and a plurality of devices are generally arranged in the factory building, so that a plurality of target points are arranged, and the inspection robot solves the problem of staged path planning, namely, the newly-reached target point is used as a new starting point, and a new path is planned again to reach the next target point.
When the robot performs global path planning, rollover situations can not be avoided in the process of passing through dangerous areas, and at the moment, the inspection work of the robot can be terminated due to rollover, so that the robot can automatically overturn to a normal state under the condition of no human intervention, and navigation is continued to finish the inspection work.
Through the above analysis, the problems and defects existing in the prior art are as follows:
(1) How the inspection robot can build a two-dimensional map of surrounding obstacles and also build a two-dimensional map of rough terrain.
(2) How the inspection robot combines laser radar data (two-dimensional) and depth camera data (three-dimensional) to build a two-dimensional environment map of the slope.
(3) When the inspection robot performs global path planning, the inspection robot not only has the characteristics of obstacle avoidance, shortest walking path, least time consumption and the like, but also needs to select proper observation distance and observation angle to determine the proper observation range of each electrical device.
(4) How to solve the problem of staged path planning of the inspection robot, namely, taking the target point which is just reached as a new starting point, and re-planning a new path to reach the next target point.
(5) How the robot automatically turns to a normal state under the condition of no human intervention, and navigation is continued to finish the inspection work.
The difficulty of solving the problems and the defects is as follows: currently, most inspection robots adopt radar technology to survey surrounding conditions, but the detection means can only be suitable for large, high and stationary or slow-moving obstacles, and cannot meet the obstacles (pedestrians) with more obstacles and fast-moving obstacles and the lower and smaller obstacles (small pits, depressions). However, if the image technology is applied to road condition recognition, the problems are certainly solved, because the vision is more comprehensive and visual compared with the laser ranging.
The water pump house environment is provided with various barriers such as a water reservoir, a deep well, a cable, falling rocks, crushed coal and the like on the ground, and the ground is raised, collapsed, torn and the like due to rock stratum movement. The ground condition is not detected by the laser radar, and the inspection robot directly passes through the road section, so that collision or rollover accidents can occur. In a complex ground environment, in order for a robot to walk autonomously and complete equipment inspection, the advance road condition information of the robot must be detected, which is a precondition for ensuring the motion safety of the robot.
Disclosure of Invention
Aiming at the problems existing in the prior art, the invention provides a terrain awareness method, a terrain awareness system, a storage medium and computer equipment.
The invention is realized in that a terrain awareness method comprises the following steps:
step one, obtaining depth information of road surface conditions by using a Kinect depth camera, establishing a Digital Elevation Model (DEM), dividing grids based on the depth information, and dividing dangerous grade values of terrain information to realize automatic identification of dangerous grades of the road surface;
step two, combining an odometer auxiliary method with a Kalman filtering algorithm, firstly, spreading the pose of the robot by using odometer data, and realizing pose error correction according to the matching of laser radar data and an environment map;
and step three, optimizing SLAM mapping and navigation algorithm in the ROS.
In the first step, the calculated topographic information is used as a main index of the path planning of the robot.
Further, in the third step, the optimization of the SLAM mapping and navigation algorithm in the ROS comprises the following specific processes:
establishing constraints among the pose, the natural features and the artificial features of the robot in a graph optimization mode, and then performing global optimization to obtain a map with global consistency; heuristic information is added into a path planning algorithm to guide the searching direction; and then, setting up a test environment design and performance evaluation standard, carrying out comparative analysis on experimental results in the aspects of drawing efficiency, navigation strategy and the like aiming at the improved algorithm and the original algorithm, and finally optimizing the ROS system and the related program of the singlechip program.
Another object of the present invention is to provide a terrain awareness system implementing the terrain awareness method, the terrain awareness system being provided with:
a main body;
rollers are arranged on two sides of the main body, and the rollers are sleeved with tracks; the laser radar and the Kinect camera are installed on the front side of the main body.
It is another object of the present invention to provide a storage medium for receiving user input, the stored computer program causing an electronic device to execute the terrain awareness method comprising the steps of:
step one, obtaining data information related to terrain parameters by using a Kinect depth camera, and realizing automatic identification of road surface dangerous grades;
step two, removing the motion distortion of the laser radar by combining an odometer auxiliary method with a Kalman filtering algorithm;
and step three, optimizing SLAM mapping and navigation algorithm in the ROS.
It is a further object of the present invention to provide a computer program product stored on a computer readable medium, comprising a computer readable program for providing a user input interface for implementing the terrain awareness method when executed on an electronic device.
By combining all the technical schemes, the invention has the advantages and positive effects that: in the invention, the distance error of the robot global positioning on the X axis and the Y axis is less than 10cm, and the angle error is less than 0.2rad; the dimension error of the grid map and the real environment is smaller than 1cm; the deviation degree of the real navigation path of the robot and the algorithm estimated path is smaller than 10cm; the mobile robot can perform loop-back when building a graph in an environment with similar structure; the variance of the point cloud data acquired by the laser radar after pretreatment is smaller than 0.1.
The invention establishes a terrain awareness system, and can identify the obstacle in the running direction of the inspection robot; establishing a motion planning model with the length of a rugged ground path and the dangerous grade of a robot passing through a road surface as indexes, and planning the shortest path between a starting point and a terminal point under the condition of ensuring the safety of a walking route of the inspection robot; calibrating the odometer by adopting an iterative least square method to reduce the accumulated error of the robot odometer, simultaneously establishing constraints among the pose, the natural features and the artificial features of the robot by using a graph optimization mode, and performing global optimization on the map to finally obtain a map with global consistency; through SLAM construction and navigation algorithm, realize colliery underground inspection robot and carry out inspection work to the pump house equipment under unmanned control, under the condition that no track, cable and GPS assist.
Drawings
Fig. 1 is a flowchart of a terrain awareness method according to an embodiment of the present invention.
Fig. 2 is a schematic structural diagram of a terrain awareness system according to an embodiment of the present invention.
In fig. 2: 1. a main body; 2. a roller; 3. a track; 4. a laser radar; 5. kinect camera.
Fig. 3 is a schematic view of ground perception of an unmanned vehicle according to an embodiment of the present invention.
Fig. 4 is a schematic diagram of ground perception of a water pump house inspection robot according to an embodiment of the present invention.
FIG. 5 is a schematic diagram of a path planning index for ROS algorithm without ground perception provided by an embodiment of the present invention.
FIG. 6 is a schematic diagram of a path planning index for the ROS algorithm in the presence of ground awareness, as provided by an embodiment of the present invention.
Fig. 7 is a schematic view of topographic information provided by an embodiment of the present invention.
Fig. 8 is a schematic diagram of constraint of a robot pose in SLAM mapping provided by an embodiment of the present invention.
Fig. 9 is a schematic diagram of an algorithm of a×shortest path search according to an embodiment of the present invention.
Fig. 10 is a technical block diagram provided by an embodiment of the present invention.
FIG. 11 is a schematic diagram of the effect of path planning in combination with topographical factors provided by an embodiment of the present invention.
Fig. 12 is a schematic diagram showing comparison of effects of the lidar provided by the embodiment of the invention before and after motion distortion removal.
Fig. 13 is a schematic diagram showing comparison of the effects of the pre-and post-optimization mapping according to the embodiment of the present invention.
Fig. 14 is a schematic diagram of a PRM path planning algorithm simulation result provided by an embodiment of the present invention.
Fig. 15 is a schematic path diagram generated by a PRM path planning algorithm according to an embodiment of the present invention.
FIG. 16 is a schematic diagram of a Pure Persuit path tracking algorithm simulation result provided by an embodiment of the invention.
FIG. 17 is a schematic diagram of a Pure Persuit algorithm trace path process analysis provided by an embodiment of the invention.
Fig. 18 is a schematic diagram of simulation results of an obstacle avoidance algorithm according to an embodiment of the present invention.
Fig. 19 is a schematic diagram illustrating a process analysis of an obstacle avoidance algorithm according to an embodiment of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the following examples in order to make the objects, technical solutions and advantages of the present invention more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
In view of the problems existing in the prior art, the present invention provides a terrain awareness method, a system, a storage medium, and a computer device, and the present invention is described in detail below with reference to the accompanying drawings.
Other steps may be performed by those of ordinary skill in the art of terrain awareness methods provided by the present invention, and the terrain awareness method provided by the present invention of fig. 1 is merely one specific example.
As shown in fig. 1, a terrain awareness method provided by an embodiment of the present invention includes:
s101: and obtaining data information related to the terrain parameters by using a Kinect depth camera, and realizing automatic identification of the road surface dangerous grade.
S102: and removing the laser radar motion distortion by combining an odometer auxiliary method with a Kalman filtering algorithm.
S103: optimization of SLAM mapping and navigation algorithms in ROS.
In S101 provided by the embodiment of the present invention, the calculated topographic information is used as a main indicator of the path planning of the robot.
In S103 provided by the embodiment of the present invention, the specific process of optimizing the SLAM mapping and navigation algorithm in ROS is:
establishing constraints among the pose, the natural features and the artificial features of the robot in a graph optimization mode, and then performing global optimization to obtain a map with global consistency; heuristic information is added into a path planning algorithm to guide the searching direction, so that the time of global path planning is reduced; and then, setting up a test environment design and performance evaluation standard, carrying out comparative analysis on experimental results in the aspects of drawing efficiency, navigation strategy and the like aiming at the improved algorithm and the original algorithm, and finally optimizing the ROS system and the related program of the singlechip program.
As shown in fig. 2, in the terrain sensing system provided by the embodiment of the invention, rollers 2 are arranged on two sides of a main body 1, and a crawler belt 3 is sleeved on the rollers 2; a laser radar 4 and a Kinect camera 5 are mounted on the front side of the main body 1.
The technical scheme of the invention is further described below with reference to specific embodiments.
1. A problem with robot navigation is a technique and method for guiding a mobile robot from a starting point to a target point. In recent years, home and abroad scholars have summarized navigation problems into three problems: (1) positioning problems; (2) identification of targets; (3) Path planning problem.
The positioning problem is used as the most basic problem of realizing autonomous capability of a mobile robot, and is the core of the robot navigation problem. Different positioning modes generally fall into the following three categories:
(1) Relative positioning. Dead-Reckoning (DR), which mainly relies on body sensors such as encoders, gyroscopes, accelerometers and the like to calculate the current moment position, attitude and other information of the mobile robot by measuring the relative distance, angular velocity, linear acceleration and other information according to initial state information;
(2) And (5) absolute positioning. The absolute positioning mainly uses positioning sensors such as RFID, GPS and the like, and the principle is that absolute position information of a robot is determined by using technical methods such as navigation beacons, map matching or satellite positioning;
(3) And (5) combined positioning. The relative positioning is performed according to the robot kinematic model and the internal sensor, the signals are not interfered by the outside, but the errors are accumulated along with time.
The development trend of mobile robot navigation mainly has the following aspects:
(1) Real-time, accurate and stable navigation modes. There are many navigation methods in common use, but each has certain limitations. Therefore, while improving the accuracy of the sensor, some intelligent algorithms are added to be applied to navigation tasks.
(2) And a plurality of navigation modes are matched for use. The single navigation technology has inherent limitations, not only perfects a single navigation mode, but also comprehensively combines a plurality of navigation modes, adopts a plurality of sensors to mutually compensate respective defects, and is more stable in navigation system and capable of completing navigation tasks in more complex and changeable environments.
(3) As network technology and wireless technology mature, remote control of robots can be achieved using network and wireless technology, which is also a development direction of mobile robot navigation technology.
(4) The modularization problem of the system. Because the navigation system can be divided into a plurality of modules with different layers, in order to achieve a certain degree of uniformity and universality, the problem of modularization of navigation and unification of interfaces of the navigation system is studied, and is also the key point of future research.
2. Content
(1) Mobile robot platform kinematics and dynamics analysis
According to the research basis of main technical indexes and parameters of the existing mining water pump house inspection robot, the following related researches are firstly carried out on the inspection robot.
1) Design wheel type mining water pump house inspection robot simulation platform
According to the size of a sensor carried on the water pump room inspection robot and the size of each part of the inspection robot, a UG three-dimensional model of the water pump room inspection robot is designed, statics finite element analysis is carried out on the whole inspection robot, and finally stress and strain diagrams of each part are obtained, so that whether the model selection of each part is reasonable or not is determined.
2) Robot platform model kinematics and dynamics analysis
The UG three-dimensional model is imported into ADAMS software, and through the kinematic analysis of the robot platform model, whether each part of the robot has conflict in the motion process is determined, and whether the model selection of the part is reasonable is further determined; the dynamic analysis of the robot platform can obtain the change of the driving force of the robot in the processes of walking, obstacle surmounting and stair climbing, so that a reference is provided for the selection of the power of the driving wheel motor of the robot.
(2) Terrain perception method
In order for the robot to better route planning, the following related studies are to be performed:
1) Collecting data information about topographical parameters
The ground of the water pump house is obviously different from the ground condition in a laboratory, and various barriers such as a reservoir, a deep well, a cable, falling rocks, crushed coals and the like exist on the ground, so that the ground is raised, collapsed, torn and the like due to rock stratum movement. If the ground bulge is smaller than the car body, the ground bulge cannot be detected by the laser radar, the inspection robot can directly pass through the road section, and collision or rollover accidents can possibly occur; if the ground is sunken, the inspection robot passes through the road section during navigation, possibly being blocked in the sunken part and not being capable of advancing. Therefore, in a complex ground environment, in order for a robot to walk autonomously and complete equipment inspection, ground condition information of the environment in which the robot is located must be considered, which is a precondition for ensuring the motion safety of the robot. Information on the topographic parameters needs to be collected, specifically analyzing the morphological features of the raised and recessed floors.
2) Method and process for automatically identifying road surface danger level by research
(1) Unmanned vehicle ground perception
The laser radar sensor for front obstacle avoidance is installed on a front bumper of a vehicle in a downward inclined mode by 15 degrees, is 1.2m away from the ground, and can be used for roadway ground in a range of 4.5m in front. It can be seen from the figure that the included angle between the whole laser scanning plane and the road surface is 15 degrees, and the CCD camera sensor for identifying the front object is arranged at the center position of the top of the vehicle, and the distance from the ground is 1.6m, and because the visual angle of the camera is 30 degrees, the camera is required to be installed in a downward tilting mode by 17 degrees for joint identification with the laser radar sensor, so that the identification point of the camera can be synchronous with the sensing point of the laser radar sensor, the front object can be better identified, the identification point corresponds to the laser radar identification road surface point, and the ground obstacle in 4.5m in front of the vehicle can be better identified, as shown in fig. 3.
(2) Ground sensing of water pump house inspection robot
And designing the ground sensing scheme of the ROS inspection robot by referring to the ground sensing scheme of the unmanned automobile. The ROS inspection robot is provided with a two-dimensional laser, the radar can not detect the ground condition like a three-dimensional laser radar, so that the two-dimensional laser radar can be loaded on the upper part of the vehicle body, the surrounding obstacles higher than the vehicle body are mapped, the front part of the vehicle body of the inspection robot is provided with a Kinect depth camera, and the condition of the ground is detected by using the Kinect depth camera, as shown in figure 4.
(3) Complex ground condition map
As shown in fig. 5, the algorithm in the general ROS does not consider the ground condition factor, and the established map only contains the obstacle information with the height greater than the ROS vehicle body height, and the reference index is only the shortest path and no obstacle when the path planning is performed. Considering that the inspection robot may pass through the road sections with raised ground and depressed ground in the process of executing the path, and cause the accident of turning over and trucks, the complex ground condition needs to be mapped.
Therefore, in the process of robot map building, the two-dimensional laser radar is responsible for building a surrounding two-dimensional obstacle map, the Kinect arranged at the front end of the ROS inspection robot is responsible for building a ground map, the trafficability of the robot to a walking area is evaluated through three indexes of flat ground, shortest path and no obstacle, the passable and non-passable areas are divided, and the optimal path is ensured to be obtained by robot path planning, as shown in fig. 5.
3) Calculation method and implementation of topographic information perception obtained by research and calculation
The measurement of the morphology of the landform is considered from three aspects of height, ground surface inclination azimuth and degree, ground plane shape and area. In the process of robot technology research, the topography is not only in a rugged state, but also takes the ground material and other contents into consideration. Therefore, the problems of obstacle avoidance, path planning and the like of the robot can be causedStudy was conducted. The representation of the topographic features comprises the contents of topographic relief, topographic texture, topographic gradient and the like, and the three contents can effectively represent the topographic information features. The terrain information is mainly represented by a digital elevation model DEM, wherein the model mainly comprises regular grids, and each grid has an average elevation corresponding to a grid area. The grid is divided based on DEM, so that it is known that the D value is significantly affected by the terrain gradient S (slip constraint), the waviness H (step height and obstacle edge constraint) and the roughness R (body stability constraint), and in general, a 3×3DEM grid is used for numerical calculation, as shown in fig. 7, where 0 To E (V) 8 For the height value of this area grid.
(3) Laser radar data preprocessing research
In order to remove the lidar motion distortion, the following related studies are to be performed:
1) High-precision and high-adaptability filtering algorithm research
During the process of acquiring two-dimensional point cloud data, the laser radar is affected by interference, so that some noise can occur during the process of acquiring the data. Therefore, in actual work, besides errors measured by the device, the device is influenced by external environment, such as the influence factors of the measured object, the quality of the surface of the obstacle and the measured object, and the like, and in addition, some local large-scale noise cannot be filtered by the same method due to the fact that the local large-scale noise is far away from the target point cloud, so that a high-precision and high-adaptability filtering algorithm is required to be selected for filtering the two-dimensional point cloud data.
2) Mileage meter auxiliary method and Kalman filtering combined algorithm research
The single chip microcomputer is used for reading laser radar data, the pose of the robot at the moment can be obtained when the laser spot data are read each time, the motion distortion is eliminated according to the pose of the robot, and after one frame of complete data is obtained, the complete data are uploaded to the processor. And the motion distortion is eliminated in the singlechip layer, the time synchronization problem is not needed to be considered, the data is required to be compressed, and otherwise, larger delay is generated.
Solving the pose of the robot corresponding to each laser point in the laser data of the current frame, namely solving {t s ，t s+Δt ，…t e And (3) converting all laser points into the same coordinate system according to the solved pose of the robot at the moment, repackaging the laser points into a frame of laser data, and publishing the frame of laser data. Let p be s And p s+1 There are N poses { p } s ，p s1 ，…，p s(n-2) ，p sn ，p s+1 And (3) then: one frame of laser data comprises n laser points, and the pose { p } corresponding to each laser point 1 ，p 2 ，…，p n Interpolation by the method described above to obtain xi as the coordinate before conversion, x' i For the coordinates after conversionConverting the converted coordinates into laser data and issuing the laser data:
(4) SLAM mapping and navigation algorithm optimization design in ROS system
1) Study on constraints and global optimization among pose, natural features and artificial features of robot to obtain map with global consistency
As shown in fig. 8, the lidar may extract features such as corner points, line segments, circular arcs, etc. as natural beacons for positioning navigation of the robot. Among the features constituted by various geometric elements, the line segment features are a simpler and easily identified type of features that can effectively describe the structured environment. After extracting natural features in laser radar scanning data, matching the features in the scanning data at adjacent moments to solve the coordinate transformation relation between the features, and then incrementally determining the position of the robot, thereby achieving the purpose of positioning the robot. Besides extracting natural features such as points, line segments and the like existing in the environment to perform data association, the front-end algorithm of the laser SLAM can also extract artificial features to perform data association. An artificial feature is some marker, also called artificial beacon, that is artificially placed in the environment. Compared with natural features, the artificial features have the characteristics of easy extraction and easy matching, and can be suitable for complex unstructured environments. When natural features are not abundant in the working environment of the robot, artificial beacons can be arranged in the environment, and the SLAM algorithm performs positioning navigation by extracting the artificial features, so that the robustness of the SLAM algorithm can be effectively improved.
2) Path planning algorithm integrated with heuristic information is researched to obtain guiding searching direction, so that time for global path planning is reduced
The cost value is evaluated to search for the nearest path by heuristic search a. For the cost value of any one point, there is f=g+h; f represents the cost to the target point, G represents the distance from the starting point to the point, H represents the distance from the point to the target point, and the distance algorithm is manhattan distance, i.e., the sum of the number of horizontal and vertical squares from the current grid to the destination grid. The principle of the shortest path search algorithm is shown in fig. 9.
3. Target object
(1) Through researching a mapping algorithm (SLAM), the inspection robot can be accurately positioned in the water pump room, and a two-dimensional grid map of the water pump room is built on the basis.
(2) By researching a path planning algorithm (PRM), the method can realize that a path passing through all pump house equipment is planned in the pump house with the two-dimensional grid map established.
(3) By researching a path tracking algorithm (Pure Persui), the robot can be controlled to move according to a planned path, and meanwhile, obstacles can be avoided in time when the obstacles suddenly appear on the planned path.
4. Key problems to be solved
(1) Problem of drawing
(1) The mining water pump house inspection robot encounters severe natural conditions and complex geological environments, for example, various barriers such as accumulated water, cables, falling rocks and crushed coal are often arranged in a roadway; the roadway surface gradient is large, and the like, so that how the inspection robot can establish a two-dimensional map for surrounding obstacles and a two-dimensional map for rugged terrain is one of the key problems to be solved by the invention.
(2) The robot can not encounter a slope road section in the walking process, and the two-dimensional laser radar is used for building a map aiming at an obstacle on a horizontal plane in a space, so the slope is regarded as an obstacle, and the robot can not pass through the slope road section, so how to integrate laser radar data (two-dimensional) and depth camera data (three-dimensional) to build a two-dimensional environment map of the slope by the inspection robot is a second key problem to be solved by the invention.
(2) Autonomous navigation problem
(1) The inspection robot is provided with a camera or an infrared thermal imager to observe the equipment, and the size of the electrical equipment in the observation field determines the quality of data acquisition precision, so that the inspection robot not only shows the characteristics of avoiding obstacles, having shortest walking path, having least time and the like when performing global path planning, but also needs to select proper observation distance and observation angle to determine the proper observation range of each electrical equipment, and is the third key problem to be solved by the invention.
(2) The invention aims to solve the fourth key problem of solving the problem of periodic path planning by taking the just-arrived target point as a new starting point and re-planning a new path to arrive at the next target point.
(3) When the robot performs global path planning, rollover situations can not be avoided in the process of passing through dangerous areas, and at the moment, the inspection work of the robot can be terminated due to rollover, so that how the robot automatically turns to a normal state under the condition of no human intervention, and navigation is continued to finish the inspection work is a fifth key problem to be solved by the invention.
5. Method, technical route, experimental protocol and feasibility analysis
5.1 method
The invention aims to carry out a research method combining theoretical analysis, algorithm design, experimental research and simulation so as to solve the key problems encountered when the positioning and navigation algorithm in the current ROS is applied to underground coal mine equipment inspection. According to the composition structure of the ROS inspection robot, the main research content is perfected according to the working flow and the real-time environment, a three-dimensional model of the ROS robot is built, the physical parameters of the model are set, the simulation of algorithms is realized by combining Matlab and Gazebo, and an optimal algorithm suitable for inspection of the underground coal mine water pump house inspection robot is found.
5.2 technical route, as shown in fig. 10;
5.3 protocol
(1) Platform kinematics and dynamics analysis of water pump house inspection robot
According to main technical indexes and parameters of the water pump house inspection robot, a wheel type coal mine water pump house inspection robot simulation platform is designed, and kinematic and dynamic analysis is carried out on the robot platform model. The method comprises three-dimensional model design, kinematic model establishment and dynamic model establishment. The three-dimensional model of the motion chassis of the water pump house inspection robot is shown in fig. 2.
(2) Study of terrain awareness method
And obtaining data information related to the terrain parameters by using a Kinect depth camera, and realizing automatic identification of the road surface dangerous grade. And taking the calculated topographic information as a main index of the path planning of the robot. The effect of path planning in combination with topographical factors is shown in fig. 11.
(3) Preprocessing of lidar data
And removing the laser radar motion distortion by combining an odometer auxiliary method with a Kalman filtering algorithm. The laser radar motion distortion removal front-to-back effect pair is shown in fig. 12, for example.
(4) Optimization of SLAM mapping and navigation algorithms in ROS
Establishing constraints among the pose, the natural features and the artificial features of the robot in a graph optimization mode, and then performing global optimization to obtain a map with global consistency; heuristic information is added into a path planning algorithm to guide the searching direction, so that the time of global path planning is reduced; and then, setting up a test environment design and performance evaluation standard, carrying out comparative analysis on experimental results in the aspects of drawing efficiency, navigation strategy and the like aiming at the improved algorithm and the original algorithm, and finally optimizing the ROS system and the related program of the singlechip program. The graph effect pairs before and after graph optimization are shown in fig. 13.
6. Feasibility analysis
6.1 feasibility of theoretical investigation
The method is characterized in that the statics, kinematics and dynamics simulation are carried out on the three-dimensional model of the built water pump house inspection robot in the aspect of a robot body, so that the motion characteristics of the inspection robot can be continuously optimized, the performance index required by the water pump house inspection robot can be achieved, in the aspect of an algorithm, the SLAM map construction and navigation algorithm of an ROS open source is adopted in the project, the algorithm is widely applied to indoor sweeping robots, and if the algorithm is used for the underground water pump house equipment inspection robot of a coal mine, the original SLAM map construction and navigation algorithm must be properly adjusted due to the change of the working environment.
Meanwhile, the design of the invention is scientific and reasonable, and the study of the drawing and navigation algorithm of the water pump house inspection robot can liberate the water pump house inspection personnel from daily simple, mechanical and repetitive work, thereby facilitating the people to concentrate on more important and complex work, and having important significance for strengthening the safety management of coal mine production and really realizing the efficiency reduction of personnel and energy conservation and consumption reduction.
6.2 feasibility of practical research
The national and local combined engineering laboratory for mine fluid control has many years of experience in industrial automation and has been equipped with a variety of electromechanical equipment test conditions. Existing equipment has NI that can be tested and measured, providing a modular hardware platform and system design software (LabVIEW). Can be used for experiments to meet the overall requirements.
7. The SLAM construction and autonomous navigation algorithm of the underground coal mine water pump house inspection robot is researched, and the main innovation points are as follows:
(1) And a terrain sensing system is established, so that obstacles, slopes, roughness and danger levels in the running direction of the inspection robot can be identified.
(2) Based on the terrain perception information of the inspection robot, a motion planning model taking the optimized rugged ground path length and the dangerous grade of the robot when the robot passes through the road surface as indexes is established, and the shortest path between the starting point and the destination is planned under the condition that the walking route of the inspection robot is ensured to be safe.
(3) The linear least square method is adopted to calibrate the odometer, the accumulated error of the robot odometer is reduced, meanwhile, the constraint is established among the pose, the natural characteristic and the artificial characteristic of the robot in a graph optimization mode, then the map is subjected to global optimization, and finally the map with global consistency can be obtained.
(4) And a two-dimensional grid map of the water pump room is established by using a laser radar through an SLAM algorithm, a path passing through all the water pump room devices is obtained in the grid map through a path planning algorithm, and finally the robot is controlled to track the path through a path tracking algorithm so as to reach the position of each water pump room device. The underground coal mine inspection robot can carry out inspection work on the water pump room equipment under the conditions of unmanned control, no track, no cable and GPS assistance.
The technical scheme of the invention is further described below in connection with simulation experiments.
1. Work foundation
The structural design of the water pump house inspection robot prototype is finished, the simulation of a path planning algorithm, a path tracking algorithm and an obstacle avoidance algorithm is finished in Matlab, and the simulation of the SLAM mapping algorithm is finished next.
1.1 PRM path planning algorithm simulation
Before path planning is carried out, firstly expanding a two-dimensional grid map of a water pump room according to the size of a robot so as to prevent the robot from colliding with an obstacle at a corner during path tracking, then giving the expanded map to a PRM path planning algorithm, setting a starting point (1.2,2.4) and an ending point (2.7,3.3) of a planned path, wherein the number of random points generated by the PRM path planning algorithm is 800, and the simulation result of the PRM path planning algorithm is shown in fig. 14. In fig. 15 it can be seen that the path is made up of a series of links of coordinate points which play an important role in the following path tracking.
1.2 Pure Persuit path tracking algorithm simulation
Before path tracking, a two-wheel differential robot model is generated in matlab, the diameter of the robot is set to be 0.2m, the robot can obtain the pose under the world coordinate system at any moment through an odometer, and the pose can judge whether the robot reaches a target point or not; the planned path from the point 4 to the point 6 is stored in a Pure Persuit path tracking algorithm, the planned path from the point 4 to the point 6 consists of a plurality of coordinate point connecting lines, the coordinate points are generated by a PRM algorithm, and the coordinate points are target points which need to be tracked by the Pure Persuit algorithm; the simulation results obtained by setting the linear velocity of the robot to 0.1m/s and the maximum angular velocity of the robot to 1rad/s, the distance threshold to 0.01m, and executing the Pure Persuit algorithm are shown in FIG. 16.
Fig. 17 shows that the Pure Perwait path tracking algorithm controls the changes of the attitude angle, the angular velocity and the linear velocity at each moment when the robot tracks the path 4 and the path 5, the robot is at the starting point in the interval 1, the pose (X, y, theta) = (1.2,2.4,0), the original path attitude angle (the included angle between the original path and the world coordinate system X-axis) is larger than the attitude angle of the robot (the included angle between the robot coordinate system X-axis and the world coordinate system X-axis), so the robot rotation angular velocity given by the Pure Perwait path tracking controller is larger than 0, the robot rotates anticlockwise, the attitude angle of the robot increases, the attitude angle of the robot is equal to the original path attitude angle in the interval 2, the robot rotation angular velocity given by the Pure Perwait path tracking controller is equal to 0, the linear velocity is 0.1m/s, and the robot moves along the original path. In the section 3, since the attitude angle of the original path is further increased, the rotation angular velocity of the robot, which is given by the Pure Persuit path tracking controller, is greater than 0, the robot rotates counterclockwise, and the attitude angle of the robot is also increased. In the interval 4, the attitude angle of the robot is equal to the attitude angle of the original path, the rotation angular speed of the robot, which is given by the Pure Persuit path tracking controller, is equal to 0, the linear speed is still unchanged, and the robot moves along the original path. Similarly, the attitude angle of the original path is changed in the interval 5, and meanwhile, the Pure Persuit path tracking controller also gives different rotation angular speeds of the robot, so that the attitude angle of the robot is finally equal to the attitude angle of the original path. In the interval 6, the robot reaches the final point of the original path, and at this time, the linear speed and the angular speed are both 0, and the final attitude angle of the robot is still equal to the attitude angle of the original path.
1.3 obstacle avoidance algorithm simulation
The robot obstacle avoidance is that when the robot performs path tracking, if the laser radar detects that an obstacle exists in front of the robot, the obstacle avoidance algorithm is operated to control the robot to bypass the obstacle, and then the original path is continuously tracked. And tracking a path 2 and a path 3 through a Pure Persuit path tracking algorithm, and setting an obstacle on the path 2 for simulating an obstacle avoidance algorithm. The simulation results are shown in fig. 18.
Fig. 19 shows the change of the distances between the robot and the obstacle at each moment in the obstacle avoidance process, wherein the distance threshold is set to 0.1m, the laser radar sensor is turned on, the distance between each direction of the robot and the obstacle can be obtained by analyzing the data of the laser radar sensor, when the distance is greater than the distance threshold, the robot is in a path tracking stage, the path tracking algorithm controls the robot to move along the original path, when the distance between the robot and the obstacle at the front or the left is smaller than the distance threshold, the robot is in the obstacle avoidance stage, starts to execute the obstacle avoidance algorithm and stops executing the path tracking algorithm, and because the distance between the robot at the right and the obstacle is obviously greater than the front and the left, the robot turns right (rotates clockwise), the robot turns left after advancing a distance, the distance between each direction of the robot and the obstacle is detected to be greater than the distance threshold, the robot is illustrated to bypass the obstacle, the robot returns to the path tracking stage, and continues to execute the path tracking algorithm.
It should be noted that the embodiments of the present invention can be realized in hardware, software, or a combination of software and hardware. The hardware portion may be implemented using dedicated logic; the software portions may be stored in a memory and executed by a suitable instruction execution system, such as a microprocessor or special purpose design hardware. Those of ordinary skill in the art will appreciate that the apparatus and methods described above may be implemented using computer executable instructions and/or embodied in processor control code, such as provided on a carrier medium such as a magnetic disk, CD or DVD-ROM, a programmable memory such as read only memory (firmware), or a data carrier such as an optical or electronic signal carrier. The device of the present invention and its modules may be implemented by hardware circuitry, such as very large scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, etc., or programmable hardware devices such as field programmable gate arrays, programmable logic devices, etc., as well as software executed by various types of processors, or by a combination of the above hardware circuitry and software, such as firmware.
The foregoing is merely illustrative of specific embodiments of the present invention, and the scope of the invention is not limited thereto, but any modifications, equivalents, improvements and alternatives falling within the spirit and principles of the present invention will be apparent to those skilled in the art within the scope of the present invention.
Claims (3)
1. A terrain awareness method, comprising:
obtaining data information related to the terrain parameters by using a Kinect depth camera, and realizing automatic identification of the road surface dangerous grade;
removing the motion distortion of the laser radar by combining an odometer auxiliary method with a Kalman filtering algorithm; solving the pose of the robot corresponding to each laser point in the laser data of the current frame, namely solving { t } s ，t 2+Δt ，…，t e The robot pose at the moment is converted into all laser points under the same coordinate system according to the solved pose, and the laser points are repackaged into a frame of laser data and released; let p be s And p s+1 There are N poses { p } s ，p s1 ，…，p s(n-2) ，p sn ，p s+1 And (3) then: one frame of laser data is provided with n laser points, eachPose { p } corresponding to laser point 1 ，p 2 ，…，p n Interpolation to obtain x i To convert the coordinates before i For the transformed coordinates, x i To convert the coordinates before' i For the coordinates after conversion, then:converting the converted coordinates into laser data and issuing the laser data;
SLAM mapping and navigation algorithm optimization in ROS;
taking the calculated topographic information as a main index of path planning of the robot;
the SLAM mapping and navigation algorithm optimization in the ROS comprises the following specific processes: establishing constraints among the pose, the natural features and the artificial features of the robot in a graph optimization mode, and then performing global optimization to obtain a map with global consistency; heuristic information is added into a path planning algorithm to guide the searching direction; and then, setting up a test environment design and performance evaluation standard, carrying out comparative analysis of experimental results on the aspect of drawing efficiency and navigation strategy aiming at the improved algorithm and the original algorithm, and finally optimizing the ROS system and the related program of the singlechip program.
2. A terrain awareness system implementing the terrain awareness method of claim 1, wherein the terrain awareness system is provided with:
a main body;
rollers are arranged on two sides of the main body, and the rollers are sleeved with tracks; the laser radar and the Kinect camera are installed on the front side of the main body.
3. A storage medium receiving user input, the stored computer program causing an electronic device to perform the terrain awareness method of claim 1 comprising the steps of:
step one, obtaining data information related to terrain parameters by using a Kinect depth camera, and realizing automatic identification of road surface dangerous grades;
step two, removing the motion distortion of the laser radar by combining an odometer auxiliary method with a Kalman filtering algorithm;
and step three, optimizing SLAM mapping and navigation algorithm in the ROS.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110710851.5A CN113465728B (en) | 2021-06-25 | 2021-06-25 | Terrain awareness method, system, storage medium and computer equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110710851.5A CN113465728B (en) | 2021-06-25 | 2021-06-25 | Terrain awareness method, system, storage medium and computer equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113465728A CN113465728A (en) | 2021-10-01 |
| CN113465728B true CN113465728B (en) | 2023-08-04 |
Family
ID=77872965
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110710851.5A Active CN113465728B (en) | 2021-06-25 | 2021-06-25 | Terrain awareness method, system, storage medium and computer equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113465728B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114004874B (en) * | 2021-12-30 | 2022-03-25 | 贝壳技术有限公司 | Acquisition method and device of occupied grid map |
| CN114863657B (en) * | 2022-05-20 | 2023-09-29 | 深圳市威宇智通科技有限公司 | Remote taking over rescue system for vehicle traffic accident based on 4G and 5G Internet of vehicles |
| CN114952867B (en) * | 2022-07-26 | 2022-10-25 | 中国工业互联网研究院 | Control method and device for industrial robot, electronic equipment and readable storage medium |
| CN116543018B (en) * | 2023-05-08 | 2023-12-19 | 重庆市二零八地质环境研究院有限公司 | High-steep dangerous rock collapse movement analysis method based on video feature point dynamic tracking |
| CN117270576B (en) * | 2023-11-22 | 2024-02-02 | 自然资源部第三地形测量队(黑龙江第二测绘工程院) | Control method and control terminal of terrain measurement unmanned aerial vehicle |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105865449A (en) * | 2016-04-01 | 2016-08-17 | 深圳杉川科技有限公司 | Laser and vision-based hybrid location method for mobile robot |
| CN107656545A (en) * | 2017-09-12 | 2018-02-02 | 武汉大学 | A kind of automatic obstacle avoiding searched and rescued towards unmanned plane field and air navigation aid |
| WO2018121448A1 (en) * | 2016-12-30 | 2018-07-05 | 深圳市杉川机器人有限公司 | Topology map creation method and navigation method for mobile robot, programmable device, and computer readable medium |
| CN109729322A (en) * | 2019-02-20 | 2019-05-07 | 重庆工程职业技术学院 | Coal mine visual monitor system and method |
| CN109798909A (en) * | 2019-02-01 | 2019-05-24 | 安徽达特智能科技有限公司 | A kind of method of global path planning |
| CN109839118A (en) * | 2017-11-24 | 2019-06-04 | 北京京东尚科信息技术有限公司 | Paths planning method, system, robot and computer readable storage medium |
| CN109900280A (en) * | 2019-03-27 | 2019-06-18 | 浙江大学 | A kind of livestock and poultry information Perception robot and map constructing method based on independent navigation |
| CN111582123A (en) * | 2020-04-29 | 2020-08-25 | 华南理工大学 | AGV positioning method based on beacon identification and visual SLAM |
| CN112347840A (en) * | 2020-08-25 | 2021-02-09 | 天津大学 | Vision sensor laser radar integrated unmanned aerial vehicle positioning and image building device and method |
| CN112518739A (en) * | 2020-10-22 | 2021-03-19 | 新兴际华集团有限公司 | Intelligent self-navigation method for reconnaissance of tracked chassis robot |
| CN112525202A (en) * | 2020-12-21 | 2021-03-19 | 北京工商大学 | SLAM positioning and navigation method and system based on multi-sensor fusion |
| CN112987763A (en) * | 2021-05-11 | 2021-06-18 | 南京理工大学紫金学院 | ROS-based intelligent trolley of autonomous navigation robot control system |
| CN112985410A (en) * | 2021-03-02 | 2021-06-18 | 哈尔滨理工大学 | Indoor robot self-map-building navigation system based on laser SLAM |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11300414B2 (en) * | 2019-09-17 | 2022-04-12 | Baidu Usa Llc | Estimated time of arrival based on history |
-
2021
- 2021-06-25 CN CN202110710851.5A patent/CN113465728B/en active Active
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105865449A (en) * | 2016-04-01 | 2016-08-17 | 深圳杉川科技有限公司 | Laser and vision-based hybrid location method for mobile robot |
| WO2018121448A1 (en) * | 2016-12-30 | 2018-07-05 | 深圳市杉川机器人有限公司 | Topology map creation method and navigation method for mobile robot, programmable device, and computer readable medium |
| CN107656545A (en) * | 2017-09-12 | 2018-02-02 | 武汉大学 | A kind of automatic obstacle avoiding searched and rescued towards unmanned plane field and air navigation aid |
| CN109839118A (en) * | 2017-11-24 | 2019-06-04 | 北京京东尚科信息技术有限公司 | Paths planning method, system, robot and computer readable storage medium |
| CN109798909A (en) * | 2019-02-01 | 2019-05-24 | 安徽达特智能科技有限公司 | A kind of method of global path planning |
| CN109729322A (en) * | 2019-02-20 | 2019-05-07 | 重庆工程职业技术学院 | Coal mine visual monitor system and method |
| CN109900280A (en) * | 2019-03-27 | 2019-06-18 | 浙江大学 | A kind of livestock and poultry information Perception robot and map constructing method based on independent navigation |
| CN111582123A (en) * | 2020-04-29 | 2020-08-25 | 华南理工大学 | AGV positioning method based on beacon identification and visual SLAM |
| CN112347840A (en) * | 2020-08-25 | 2021-02-09 | 天津大学 | Vision sensor laser radar integrated unmanned aerial vehicle positioning and image building device and method |
| CN112518739A (en) * | 2020-10-22 | 2021-03-19 | 新兴际华集团有限公司 | Intelligent self-navigation method for reconnaissance of tracked chassis robot |
| CN112525202A (en) * | 2020-12-21 | 2021-03-19 | 北京工商大学 | SLAM positioning and navigation method and system based on multi-sensor fusion |
| CN112985410A (en) * | 2021-03-02 | 2021-06-18 | 哈尔滨理工大学 | Indoor robot self-map-building navigation system based on laser SLAM |
| CN112987763A (en) * | 2021-05-11 | 2021-06-18 | 南京理工大学紫金学院 | ROS-based intelligent trolley of autonomous navigation robot control system |
Non-Patent Citations (1)
| Title |
|---|
| Toward Autonomous Bridge Inspection: A framework and experimental results;Jung, Sungwook;2019 16TH INTERNATIONAL CONFERENCE ON UBIQUITOUS ROBOTS (UR);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113465728A (en) | 2021-10-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113465728B (en) | Terrain awareness method, system, storage medium and computer equipment | |
| CN109059942B (en) | Underground high-precision navigation map construction system and method | |
| CN113485325A (en) | SLAM mapping and autonomous navigation method for underground coal mine water pump house inspection robot | |
| CN107389125B (en) | Underground space autonomous positioning multi-sensor intelligent detection robot | |
| Thrun et al. | Autonomous exploration and mapping of abandoned mines | |
| CN113867334B (en) | Unmanned path planning method and system for mobile machinery | |
| CN109917786A (en) | A kind of robot tracking control and system operation method towards complex environment operation | |
| CN104881027A (en) | Autonomic barrier-crossing system for wheel-track transformer station inspection robot and control method thereof | |
| CN101948011A (en) | Hexapod universal walking multifunctional moonshot robot | |
| CN202216696U (en) | Coal mine disaster relief robot navigation device based on information integration | |
| CN105128975A (en) | Six-foot bionic robot and method for optimizing leg structure with high load ratio | |
| CN110470283A (en) | Intelligent measure robot system and measurement method based on BIM | |
| CN103400416B (en) | A kind of urban environment robot navigation method based on probability multilayer landform | |
| CN205950750U (en) | Transformer station inspection robot control system that navigates based on inertial navigation | |
| CN209634595U (en) | A kind of novel mine detection robot with automatic obstacle avoiding function | |
| CN204557216U (en) | Wheel-track combined Intelligent Mobile Robot active obstacle system | |
| CN113487583A (en) | Underground roadway surface deformation detection system based on 3D point cloud slice | |
| CN116560357A (en) | Tunnel inspection robot system based on SLAM and inspection control method | |
| Dandurand et al. | All-weather autonomous inspection robot for electrical substations | |
| Androulakis et al. | Navigation system for a semi-autonomous shuttle car in room and pillar coal mines based on 2D LiDAR scanners | |
| CN204331465U (en) | A kind of machine system carrying out complex environment detection | |
| CN207059672U (en) | A kind of mobile robot for installing suspended shock absorbing mechanism additional | |
| Bakambu et al. | Field trial results of planetary rover visual motion estimation in Mars analogue terrain | |
| Rachkov et al. | Reconfigurable autonomous wheel-tracked robot | |
| CN116352722A (en) | Multi-sensor fused mine inspection rescue robot and control method thereof |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |