CN113093756A - Indoor navigation robot based on laser SLAM under raspberry group platform - Google Patents
Indoor navigation robot based on laser SLAM under raspberry group platform Download PDFInfo
- Publication number
- CN113093756A CN113093756A CN202110370120.0A CN202110370120A CN113093756A CN 113093756 A CN113093756 A CN 113093756A CN 202110370120 A CN202110370120 A CN 202110370120A CN 113093756 A CN113093756 A CN 113093756A
- Authority
- CN
- China
- Prior art keywords
- robot
- raspberry
- algorithm
- ros
- laser
- 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.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0231—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means
- G05D1/0238—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means using obstacle or wall sensors
- G05D1/024—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means using obstacle or wall sensors in combination with a laser
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0212—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory
- G05D1/0214—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory in accordance with safety or protection criteria, e.g. avoiding hazardous areas
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0212—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory
- G05D1/0221—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory involving a learning process
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0231—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means
- G05D1/0246—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means using a video camera in combination with image processing means
- G05D1/0251—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means using a video camera in combination with image processing means extracting 3D information from a plurality of images taken from different locations, e.g. stereo vision
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0259—Control of position or course in two dimensions specially adapted to land vehicles using magnetic or electromagnetic means
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0259—Control of position or course in two dimensions specially adapted to land vehicles using magnetic or electromagnetic means
- G05D1/0261—Control of position or course in two dimensions specially adapted to land vehicles using magnetic or electromagnetic means using magnetic plots
Abstract
The invention provides an indoor navigation robot based on laser SLAM under a raspberry pi platform, which is characterized by comprising: the system comprises an upper computer ROS system and a lower computer chassis control module; the ROS system of the upper computer controls the laser radar to collect environment information, a two-dimensional grid map is built, and path planning is carried out on the basis of the built map to control the robot to reach a target interest point; and the lower computer chassis control module is used for driving the two wheels to move in a differential manner, carrying out speed acquisition feedback and specified speed control, and realizing motion control by combining a PID algorithm. The method depends on a raspberry group platform, adopts a laser radar as a distance sensor, and completes construction of an indoor map, real-time positioning of an indoor robot and path navigation according to the constructed map on an ROS system of the raspberry group by using a Gmapping algorithm.
Description
Technical Field
The invention belongs to the technical field of robots, and particularly relates to an indoor navigation robot based on a laser SLAM under a raspberry pi platform.
Background
The patent with publication number CN205889192U proposes one and utilizes ultrasonic ranging module to acquire the distance of robot and testee, and magnetic force acquires robot and earth magnetism contained angle, and the input pulse of robot motor is acquireed to the movement distance to the realization can be at indoor navigation's humanoid robot.
Patent publication No. CN111578939A proposes a robot tight-combination navigation method and system considering random variation of sampling period, which performs path navigation on a robot by using the collected data of the surrounding as a state vector and the collected data of a laser spot radar and the calculated trajectory data as direction expansion.
The prior art has the following defects:
disadvantage 1: patent publication No. CN205889192U discloses that the distance is calculated by using the geomagnetic angle when drawing an indoor map, but the distance between the whole body and the surrounding of the robot is not considered, and the robot cannot respond to the drastic change of the surrounding environment.
And (2) disadvantage: the patent publication CN111578939A discloses that the solution is to navigate the robot by measuring the ambient data one time and another time, but it is too cumbersome to measure the ambient data one time and another time in a stable indoor environment.
Disclosure of Invention
Aiming at the defects and shortcomings in the prior art, the invention provides a laser SLAM-based indoor navigation robot scheme under a raspberry dispatching platform, so as to achieve the functions of indoor two-dimensional laser radar mapping, indoor plane two-dimensional information drawing, real-time positioning and target interest point navigation. The robot comprises an upper computer, namely an ROS system, and a lower computer, namely a chassis control module. The upper computer is an ROS robot system based on a raspberry platform, and the ROS robot system mainly has the tasks of controlling a laser radar to collect environmental information, constructing a two-dimensional grid map by using the environmental information, and planning a path on the basis of the constructed map by using an AI algorithm to control the robot to reach a target interest point. The lower computer is mainly completed by an stm32 development board and is responsible for driving the two-wheel differential robot to carry out speed acquisition feedback and designated speed control, and accurate motion control is realized by combining a PID algorithm. Raspberry group and chassis drive pass through serial ports communication, and the chassis is responsible for gathering and control robot motor rotation speed, and the rethread serial ports is sent for the host computer, and the host computer converts the data conversion that laser radar gathered into corresponding two-dimensional grid map in the ROS system, handles speed simultaneously to the control of speed is carried out to the lower computer with expecting speed feedback.
The robot has two wheels driven by coding motors, can scan barriers such as indoor buildings, sofa furniture and the like through Lidar, traces the information into a two-dimensional grid map through algorithms, can realize navigation of target interest places through the map, can realize some things that human beings cannot finish or are dangerous based on the technology, and has practical effect in the fields of commerce and military industry.
The invention specifically adopts the following technical scheme:
the utility model provides an indoor navigation robot based on laser SLAM under raspberry group platform which characterized in that includes: the system comprises an upper computer ROS system and a lower computer chassis control module; the ROS system of the upper computer controls the laser radar to collect environment information, a two-dimensional grid map is built, and path planning is carried out on the basis of the built map to control the robot to reach a target interest point; and the lower computer chassis control module is used for driving the two wheels to move in a differential manner, carrying out speed acquisition feedback and specified speed control, and realizing motion control by combining a PID algorithm.
Preferably, the method comprises the following steps: with wheels driven by two coded motors; adopting a raspberry pie and a PC as a host and a slave of the ROS system respectively; the driving module, the IMU module and the laser radar are respectively connected with the raspberry pie; the driving module comprises a coding motor with a code disc.
Preferably, the raspberry pi is used as a host of the ROS robot operating system for collecting data of the laser radar, and the data are visually operated by using the PC, and the raspberry pi and the PC are connected through SSH.
Preferably, the driving module communicates with the ROS robot operating system through a serial port; the adopted motor models are as follows: 25 GA-370; the adopted motor encoder is an incremental encoder; the laser radar adopted is of the type: LDS-01; the adopted IMU modules have the following types: and an MPU 6050.
Preferably, the robot is positioned by adopting a self-adaptive Monte Carlo positioning method; adopting a Gmapping algorithm as a graph building algorithm; and adopting an A-algorithm as a navigation algorithm.
Compared with the prior art, the method and the preferable scheme thereof rely on a raspberry group platform, adopt the laser radar as a distance sensor, and finish the construction of an indoor map, the real-time positioning of an indoor robot and the path navigation according to the constructed map on the ROS system of the raspberry group by utilizing the Gmapping algorithm. The map established by the radar can completely depict the whole indoor environment, although the used radar has low precision, the indoor obstacles such as sofa walls and the like can be well depicted, and the navigation precision is within the tolerable error.
The method has the advantages that:
a robot motion chassis is built, and PID (proportion integration differentiation) parameters are set, so that the left wheel and the right wheel have better response;
the method has the advantages that:
the raspberry pie is provided with an ROS system and can communicate with the chassis through a serial port;
the method has the advantages that:
the raspberry group can read information through the installed Lidar, data processing is carried out through the ROS system, and an indoor map can be constructed through the Gmapping algorithm in cooperation with a milemeter;
the advantages are that:
and a target place can be set according to the constructed map, and the robot plans a path through an A-star algorithm so as to complete indoor navigation.
Drawings
The invention is described in further detail below with reference to the following figures and detailed description:
FIG. 1 is a schematic diagram of the overall system structure according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of a kinematic model derivation process of a two-wheeled differential mobile robot according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of an adaptive Monte Carlo location process according to an embodiment of the present invention;
fig. 4 is a schematic diagram of a system a algorithm according to an embodiment of the present invention;
FIG. 5 is a schematic diagram of a relevant node for starting a robot according to an embodiment of the present invention;
FIG. 6 is a diagram illustrating an AMCL particle convergence process according to an embodiment of the invention;
fig. 7 is a schematic diagram of starting a gmaping related node according to an embodiment of the present invention;
fig. 8 is a schematic diagram of an indoor map built by using a mapping algorithm of gmaping in the embodiment of the present invention;
FIG. 9 is a schematic diagram of nodes on which graph construction depends according to an embodiment of the present invention;
FIG. 10 is a diagram illustrating a node associated with booting according to an embodiment of the present invention;
FIG. 11 is a diagram illustrating a navigation process according to an embodiment of the present invention.
Detailed Description
In order to make the features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in detail as follows:
the indoor navigation robot based on the laser SLAM under the raspberry pi platform provided by the embodiment adopts two parts of hardware, namely, the raspberry pi and the notebook computer, which are respectively used as the host and the slave of the ROS system, and the development mode of the master and the slave can improve the development efficiency, wherein the raspberry pi end is used as the host of the ROS robot operating system, and the host is responsible for acquiring data such as laser radar and the like, and the host uses the Ubuntu 16.04 system on the notebook. To perform visualization operation, and the two are connected through SSH. The protocol is shown in figure 1.
In this embodiment, the chassis driver board (driver module) and the ROS communicate via serial ports, which may also be understood as communication between STM32 and raspberry pi, or may also be considered as communication between STM32 and Ubuntu. The communication code needs to be packaged by using an ROS software framework, the communication protocol between the STM32 and the ROS needs to consider the transmission speed, size, data content and other problems, then uniform packaging is carried out, and finally a serial port sending function is called to send out the packaged data according to the byte order. In the definition of the data communication protocol, one unit type is used to define two domains with the same size, one is a buffer area data buffer for data transmission and reception, and the other is a data domain structure. The union type will only allocate memory space of the largest domain size. When the two domains are the same in size (the memory is agreed to be aligned to be 1 byte at the same time), the transparent transmission process of the data can be completed only by operating, receiving and sending the cache data through the serial port.
In this embodiment, a serial port node is designed under the ROS platform, the node subscribes a command theme sent by the control node, the command is sent to the mobile base through the serial port device, meanwhile, the serial port node receives real-time sensor data sent by the mobile base through a serial port in real time, and the data is packaged and then issued in a sensor theme mode, and the istenner node can subscribe the theme. Therefore, the serial port communication process of the ROS and the mobile base is realized.
The scheme drives two tires to be coding motors with code discs, the model of the motor is 25GA-370, the power of the motor is 12W, the rotating speed is 280RPM, the type of the motor is a direct current brush and comprises a main magnetic pole and an electric brush, a commutator and an armature winding form a rotor of the motor, direct current enters the armature winding through the commutator electric brush to generate current, the current can generate a magnetic field, the magnetic field interacts with the main magnetic field to generate electromagnetic torque, and the motor rotates accordingly. The encoder is responsible for gathering current pulse signal, and encode it, can become the format that more convenient storage and transmission were stored into the data format, the encoder can convert the displacement into the signal of telecommunication, call as the code wheel, the motor encoder that this embodiment adopted is incremental encoder, thereby can convert the odometer that the robot motion begins into unified unit such as meter per second, incremental encoder sends pulse signal through the angle of motor rotation and usually has A term B item and Z item, thereby the positive and negative direction of motor can be judged to the speed that the pulse that the output was passed through AB term arrived.
The radar transmitting frequency used by the scheme is 1.8khz, the emitted laser can be reflected after encountering an obstacle according to an optical principle, the receiver receives the reflected signal, the related module can calculate the distance through time difference, and the two-dimensional plane can be scanned by rotating according to fixed frequency, so that planar information can be generated, and the construction of a 2D map can be realized. Common distance measurement methods of the laser radar include a pulse method, a dry method and a triangle method, and the radar used in the embodiment adopts the principle of the triangle method for distance measurement. The incident light and the reflected light form a triangle, and are therefore referred to as laser triangulation. Its advantage is low requirement to hardware. Specifically, the LDS-01 radar is adopted, the 360 laser distance sensor LDS-01 is a 2D laser scanner, can sense 360 degrees, can collect a group of data around the robot for SLAM (simultaneous localization and mapping) and navigation, and the LDS-01 is used for TurtleBot3 blower, Waffle and WafflePi models. It supports a USB interface (USB2LDS), is easy to install on a PC, and supports a UART interface for an embedded development board.
The IMU is an inertial measurement unit, which consists of an accelerometer, a gyroscope, etc., and is used to measure the velocity and acceleration of three axes in space. The acceleration of gravity of three axes is measured by an accelerometer, the angular velocity is measured by a gyroscope, and the posture of an object in a three-dimensional space can be calculated by using the data, so that the method has very important use value in an SLAM navigation system. The triaxial accelerometer is greatly influenced by external force acceleration, if oscillation occurs in the motion process, the output error is also large, meanwhile, the intensity is low and the interference of other magnetic objects is easy to occur due to the fact that the intensity is taken as a reference object according to a geomagnetic field, but the data can be more stable after the gyroscope fused with the Z axis is fused, the accumulated error of the acceleration measured in the vertical downward gravity direction is small, but the defect is that the micro deformation is caused due to the fact that the gravity is the same as the gravity of the robot due to the fact that the gravity and the gravity cannot be distinguished, the error is generated when the robot does variable-speed motion in a three-dimensional space, the angle needs to be calculated by utilizing the integral of the angular velocity and the time, and then the calculated angle is summed with the initial quantity to obtain the target angle. The IMU module of this embodiment employs an MPU6050, which includes a three-axis accelerometer and a gyroscope, and these data are read through an IIC bus.
Considering that the robot system has a plurality of coordinate systems which are usually three-dimensional, the position of each coordinate system changes along with the change of the motion of the robot, and the robot designed in this time comprises a global coordinate system, a robot chassis center coordinate system and coordinate systems of each part of the robot. In these coordinate systems, the coordinate system of the robot chassis and the relative positions of its various components and sensors are not changed, but their positions in the global coordinate system are changed along with the displacement motion of the robot, which is called dynamic coordinate system. The mutual transformation between coordinate systems is important, and particularly in the process of drawing, positioning and navigation of the robot, the position judged by the laser radar according to scanning and the relative position of each part need to be accurately positioned. The ROS provided TF tool can greatly simplify this operation. tf may track a plurality of coordinate systems of the robot, which are maintained and stored by a tree-shaped data structure, called tftree, by which we can transform data of the coordinate systems between coordinate values at any time.
In a specific algorithm control flow:
the robot motion model adopted in the embodiment is a two-wheel differential mobile robot motion model, although the model is simple, the kinematic equation of the model is easy to derive and is easy to realize on a physical robot body, as shown in fig. 2, the model drives the motion of the robot by using two differential wheels, the motion direction of the robot can be controlled by using different speeds of the two wheels, and the universal steering wheel is used for controlling the balance of a vehicle body, so that the cost is saved and the precision is higher.
The kinematic model is derived as follows: wherein the relative positions of the robots in two adjacent positions, v, are shown in fig. 2LAnd vRRespectively the real-time speed of the left and right wheels, L representing the distance between the two axes of the wheel, theta1Is the angle of the two-wheeled robot moving around the circular arc, theta3Is the angle of deflection of the advancing direction of the two-wheeled robot, r isThe arc radius of this arc motion, d is the distance that one wheel advances more than the other wheel, and assuming that the two-wheeled robot is rotated to the left at this time, the right wheel moves more than the left wheel by d, so that the speed v of the robot at this time can be calculated (v ═ vL+vR) 2, while theta1＝θ2＝θ3That is, the angular rotation amount of the advancing direction of the two-wheeled robot is equal to the rotational angle of the circular arc track of the overall motion, and according to the geometrical principle, when two sections of time are close to each other, sin θ ═ θ, that is, θ can be deduced2＝sinθ2＝d/L＝(v2-vR) Δ t/L, so that the angular velocity ω θ of the two-wheeled robot moving around the center of the circle can be obtained1/Δt＝(vr-vL)/L。
Because of interference factors in a real environment and load change of the robot, it is realistic to directly use PWM to correspond to the speed of the motor, because PWM controls the current of the motor, the current is a factor determining the torque of the motor, and the current also changes correspondingly with the change of the load, directly inquiring the current corresponding to different loads is an inefficient method, while PID just provides a general method for controlling the motor, does not need to pay attention to the load of the robot, does not need to pay attention to the influence of the environment on the robot, such as factors of ground slipping and the like, and can automatically adjust within a certain range.
The robot positioning algorithm adopted in the embodiment: the AMCL is also known as adaptive Monte Carlo localization, the English name of adaptive Monte Carlo localization[The method for updating the particles is adaptive KLD, and the MCL is positioned by using a particle filtering method. The particle filtering is to compare the simulated information of each particle after the particle moves along with the robot with the actual information collected by the laser radar sensor, so as to calculate the probability of each particle, and then to regenerate the particle according to the probability, the higher probability represents the higher generation rate, the probability information is iterated continuously after the robot moves continuously, so as to converge, the position information of the robot is more and more accurate, and the whole process of the AMCL is shown in fig. 3.
And (3) a mapping algorithm: the mapping comprises a position part and a mapping part, the problem model is that the robot is placed in an unknown environment, the robot is moved, sensors such as Lidar, a depth camera and the like are used for collecting environment information, the robot is enabled to move continuously through a reasonable algorithm to collect information, a complete map of the environment is drawn step by step, and positioning and mapping are bound together and mutually influence each other.
The mapping algorithm adopted in this embodiment is an SLAM algorithm, which is a filtering-based algorithm and is an algorithm that separates two processes of mapping and positioning, that is, positioning is performed first, and then mapping is performed. The Gmapping algorithm has very good performance and high precision for a small map, so that the Gmapping algorithm is very effective as a construction algorithm of an indoor map. The laser radar system integrates pose information provided by the odometer, effectively utilizes the pose information, and therefore, the laser radar system has low frequency requirement on the laser radar. Of course, the gmaping algorithm trades space for time, and therefore becomes less efficient for large maps.
The navigation algorithm adopted in the embodiment is as follows: an algorithm A in the field of artificial intelligence is introduced into the optimal path search of the vector map, the algorithm A is a path searching algorithm and is a dijstra improved algorithm, and compared with the dijstra algorithm, the performance of the algorithm is better and the accuracy is higher.
A rough procedure of the algorithm:
(1) the selected node is placed in a list, which is open state.
(2) If the open list is empty, the description fails to match. If the target node is in the list, the success is indicated.
(3) The point in open where F is the smallest is selected and added to another list, which is the close state.
(4) And counting the adjacent nodes of the current node, and judging whether the adjacent nodes can reach or not to make the adjacent nodes be child nodes. For each child node: if it is a close state, it is discarded; if the current node is in the open state, checking whether the F value calculated by the current node is smaller, if so, updating the F value of the current node, and setting the parent node of the current node as the current node(ii) a If the node is not in the open state, adding the node into the open state, calculating the F value, and setting the parent node of the node as the current node[12]。
(5) Go to step 2
Taking fig. 4 as an example F, the cost from the starting point of the left light square to the designated square is represented by various algorithms, such as manhattan distance, euclidean distance, and the like, and G represents the distance from the designated node to the end point of the right light square, and also directly represents the manhattan distance or the euclidean distance between two points, so that the end point can be approached as much as possible, and the process of path search is ensured to be a trend toward the end point, rather than no directionality as in the dijstra algorithm.
The embodiment provides a two-dimensional laser-based SLAM robot, wherein the SLAM mainly comprises two links of map construction and positioning, and the current mainstream SLAM algorithm comprises a visual SLAM and a laser SLAM, and also comprises an SLAM combining Lidar and vision. For the laser sensor, the environmental distance information is obtained according to reflection, so that direct relative positioning is realized, and absolute positioning and track optimization for the laser sensor can be performed on the basis of relative positioning. For a visual sensor, direct distance information relative to the environment cannot be acquired, and relative change of the position of the visual sensor is required to be evaluated through deep learning by adjacent frames or a plurality of images.
Testing of ACML on machine: first, the robot starts the nodes shown in fig. 5, which are the initialized node of the robot, the keyboard control node, the amcl test node, and the camera node.
ROS starts from the slave machine rviz to visualize the AMCL particles, as shown in FIG. 6, the process of AMCL particle convergence is shown, and in the process of continuously moving the robot, the particles can be seen to gradually converge from dispersion, and finally converge to a relatively saturated region.
Gmapping is carried out for demonstrating: the Gmapping graph is built depending on two tf, respectively tf between base _ footprint and laser, which is the direct transformation of the chassis of the robot and the laser radar, and the tf is provided by static tf transformation in a robot start file robot _ start.launch; another tf is the tf between base _ footprint and odom _ combined, which provides the coordinate system of the original location of the odometer while also depending on the lidar data, of the type sensor _ msgs/LaserScan, corresponding to the transformation between the chassis and the starting point of the odometer. The starting related nodes are a robot initialization node, a keyboard control node, a camera starting node and a Gnaping mapping node as shown in FIG. 7, after the starting related nodes are started, the Ubuntu slave computer opens a rviz tool to operate a keyboard to control a robot trolley to move back and forth, a map can be built, after the robot is operated by the keyboard to circle a circle, the map is roughly built, the finally completed map is shown in FIG. 8, and the input command ROStopiclist can see the topic depending on the mapping in this time as shown in FIG. 9. After the drawing is built, the method is effective in identifying the straight wall, and can well judge the obstacles for furniture such as sofas and the like, but due to the problem of low radar precision, the resolution of a plurality of corners is not high, and meanwhile, the left room wholly has slight deviation, which is related to the reason that the floor is not flat and the vehicle slips. At this point, the mapping of Gmapping is completed, and the map can be saved to the position below the ROS space through a command and can be navigated by using the mapped map.
Navigation demonstration: firstly, the Ubuntu host is remotely connected with the raspberry dispatching robot through the SSH, and then the starting nodes shown in fig. 10, namely a robot initialization node, a navigation node and an Ubuntu slave rivz visualization map node, are started.
The process of navigation can be seen on Ubuntu through RVIZ as shown in fig. 11, the dark lines are the global navigation route established by using the a-x algorithm, the green lines are the local navigation route planned according to the DWA sliding window, and meanwhile, it can be observed that the AMCL particles are also in continuous convergence, and the convergence is gradually saturated from the initial relative dispersion.
The patent is not limited to the above preferred embodiments, and any other indoor navigation robot based on laser SLAM under raspberry platform in various forms can be derived from the teaching of the present patent, and all equivalent changes and modifications made according to the claimed invention shall fall within the scope of the present patent.
Claims (5)
1. The utility model provides an indoor navigation robot based on laser SLAM under raspberry group platform which characterized in that includes: the system comprises an upper computer ROS system and a lower computer chassis control module; the ROS system of the upper computer controls the laser radar to collect environment information, a two-dimensional grid map is built, and path planning is carried out on the basis of the built map to control the robot to reach a target interest point; and the lower computer chassis control module is used for driving the two wheels to move in a differential manner, carrying out speed acquisition feedback and specified speed control, and realizing motion control by combining a PID algorithm.
2. The indoor navigation robot based on laser SLAM under the raspberry pi platform of claim 1, characterized in that: with wheels driven by two coded motors; adopting a raspberry pie and a PC as a host and a slave of the ROS system respectively; the driving module, the IMU module and the laser radar are respectively connected with the raspberry pie; the driving module comprises a coding motor with a code disc.
3. The indoor navigation robot based on laser SLAM under raspberry pi platform of claim 2, characterized by: the raspberry pi is used as a host of the ROS robot operating system for collecting data of the laser radar, and the PC is used for performing visual operation, and the raspberry pi and the ROS robot operating system are connected through SSH.
4. The indoor navigation robot based on laser SLAM under raspberry pi platform of claim 3, characterized by: the driving module is communicated with the ROS robot operating system through a serial port; the adopted motor models are as follows: 25 GA-370; the adopted motor encoder is an incremental encoder; the laser radar adopted is of the type: LDS-01; the adopted IMU modules have the following types: and an MPU 6050.
5. The indoor navigation robot based on laser SLAM under raspberry pi platform of claim 2, characterized by: positioning the robot by adopting a self-adaptive Monte Carlo positioning method; adopting a Gmapping algorithm as a graph building algorithm; and adopting an A-algorithm as a navigation algorithm.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110370120.0A CN113093756A (en) | 2021-04-07 | 2021-04-07 | Indoor navigation robot based on laser SLAM under raspberry group platform |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110370120.0A CN113093756A (en) | 2021-04-07 | 2021-04-07 | Indoor navigation robot based on laser SLAM under raspberry group platform |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113093756A true CN113093756A (en) | 2021-07-09 |
Family
ID=76674280
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110370120.0A Pending CN113093756A (en) | 2021-04-07 | 2021-04-07 | Indoor navigation robot based on laser SLAM under raspberry group platform |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113093756A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113884574A (en) * | 2021-11-03 | 2022-01-04 | 河南理工大学 | SLAM-based intelligent pavement longitudinal section flatness measuring instrument |
| CN114089753A (en) * | 2021-11-11 | 2022-02-25 | 江苏科技大学 | Night astronomical assistant observation method based on wheeled robot |
| CN114265417A (en) * | 2022-03-01 | 2022-04-01 | 博学宽行(成都)科技有限公司 | Robot control system based on laser and visual identification navigation |
| CN115933634A (en) * | 2022-10-12 | 2023-04-07 | 海南大学 | Unknown environment exploration method, unknown environment exploration system, mobile robot and storage medium |
| CN116166027A (en) * | 2023-02-28 | 2023-05-26 | 安徽常云科技服务有限公司 | Intelligent robot control method and system for warehouse logistics |
-
2021
- 2021-04-07 CN CN202110370120.0A patent/CN113093756A/en active Pending
Non-Patent Citations (4)
| Title |
|---|
| EEFOCUS: "STM32和ROS机器人的串口通信方案", 《HTTP://NEWS.EEWORLD.COM.CN/MCU/IC517908.HTML》 * |
| 王岸雄: "基于ROS的自主移动机器人环境建模和路径规划研究", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
| 王方: "基于ROS的多履带全向移动机器人设计与实现", 《中国优秀硕士学位论文全文数据库》 * |
| 王林荣 等: "基于ROS的激光SLAM室内建图定位导航智能机器人设计", 《无线互联科技》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113884574A (en) * | 2021-11-03 | 2022-01-04 | 河南理工大学 | SLAM-based intelligent pavement longitudinal section flatness measuring instrument |
| CN113884574B (en) * | 2021-11-03 | 2024-03-19 | 河南理工大学 | Intelligent pavement longitudinal section flatness measuring instrument based on SLAM |
| CN114089753A (en) * | 2021-11-11 | 2022-02-25 | 江苏科技大学 | Night astronomical assistant observation method based on wheeled robot |
| CN114265417A (en) * | 2022-03-01 | 2022-04-01 | 博学宽行(成都)科技有限公司 | Robot control system based on laser and visual identification navigation |
| CN115933634A (en) * | 2022-10-12 | 2023-04-07 | 海南大学 | Unknown environment exploration method, unknown environment exploration system, mobile robot and storage medium |
| CN116166027A (en) * | 2023-02-28 | 2023-05-26 | 安徽常云科技服务有限公司 | Intelligent robot control method and system for warehouse logistics |
| CN116166027B (en) * | 2023-02-28 | 2023-12-26 | 湛江诚通物流有限公司 | Intelligent robot control method and system for warehouse logistics |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113093756A (en) | Indoor navigation robot based on laser SLAM under raspberry group platform | |
| WO2021254367A1 (en) | Robot system and positioning navigation method | |
| JP6868028B2 (en) | Autonomous positioning navigation equipment, positioning navigation method and autonomous positioning navigation system | |
| CN107671857B (en) | Three-dimensional simulation platform for operation demonstration and algorithm verification of service robot | |
| CN111308490B (en) | Balance car indoor positioning and navigation system based on single-line laser radar | |
| Cheng et al. | Mobile robot navigation based on lidar | |
| Li et al. | Localization and navigation for indoor mobile robot based on ROS | |
| CN110531640A (en) | A kind of comprehensive simulating method and system of robot | |
| CN113189613B (en) | Robot positioning method based on particle filtering | |
| CN108759822A (en) | A kind of mobile robot 3D positioning systems | |
| CN108646759B (en) | Intelligent detachable mobile robot system based on stereoscopic vision and control method | |
| CN109471123A (en) | Remote probe robot | |
| CN117252011A (en) | Heterogeneous ground-air unmanned cluster simulation system construction method based on distributed architecture | |
| CN109489666B (en) | Method for synchronous positioning and map construction of greenhouse pesticide spraying robot | |
| Son et al. | The practice of mapping-based navigation system for indoor robot with RPLIDAR and Raspberry Pi | |
| CN114839990A (en) | Cluster robot experiment platform | |
| Wang et al. | Research on localization and path planning of indoor robot based on ROS | |
| Marques et al. | Autonomous robot for mapping using ultrasonic sensors | |
| Rui et al. | Design and implementation of tour guide robot for red education base | |
| Hanz et al. | An abstraction layer for controlling heterogeneous mobile cyber-physical systems | |
| Mohan et al. | A comprehensive review of SLAM techniques | |
| Caprari et al. | Robot navigation in centimeter range labyrinths | |
| Morris et al. | CityFlyer: Progress toward autonomous MAV navigation and 3D mapping | |
| Mohan et al. | 22 A Comprehensive | |
| CN115145292B (en) | Terrain detection method based on wheel-foot robot joint motion analysis |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210709 |