CN116309907B - Mobile robot mapping method based on improved Gmapping algorithm - Google Patents
Mobile robot mapping method based on improved Gmapping algorithm Download PDFInfo
- Publication number
- CN116309907B CN116309907B CN202310203297.0A CN202310203297A CN116309907B CN 116309907 B CN116309907 B CN 116309907B CN 202310203297 A CN202310203297 A CN 202310203297A CN 116309907 B CN116309907 B CN 116309907B
- Authority
- CN
- China
- Prior art keywords
- particles
- particle
- map
- mobile robot
- information
- 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
- 238000000034 method Methods 0.000 title claims abstract description 29
- 238000013507 mapping Methods 0.000 title claims abstract description 16
- 239000002245 particle Substances 0.000 claims abstract description 126
- 238000012952 Resampling Methods 0.000 claims abstract description 13
- 230000015556 catabolic process Effects 0.000 claims abstract description 7
- 238000006731 degradation reaction Methods 0.000 claims abstract description 7
- 230000036544 posture Effects 0.000 claims description 16
- 238000012545 processing Methods 0.000 claims description 4
- 238000010606 normalization Methods 0.000 claims description 2
- 238000004088 simulation Methods 0.000 abstract description 18
- 230000000694 effects Effects 0.000 abstract description 6
- 230000007547 defect Effects 0.000 abstract 2
- 238000010276 construction Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000011192 particle characterization Methods 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/38—Electronic maps specially adapted for navigation; Updating thereof
- G01C21/3804—Creation or updating of map data
- G01C21/3833—Creation or updating of map data characterised by the source of data
- G01C21/3841—Data obtained from two or more sources, e.g. probe vehicles
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T11/00—2D [Two Dimensional] image generation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Automation & Control Theory (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
Abstract
The invention provides a mobile robot mapping method based on an improved Gmapping algorithm. In the process of mobile robot mapping, gmapping algorithm has good mapping effect in indoor environment, but has the defect of particle degradation, and the main defect of the invention is improved. After the laser radar carried by the mobile robot body emits particles, the description capability of the particles to the environment is changed due to the change of the two-dimensional environment. In Gmapping algorithm, a link for judging the state of the particles is added, and a resampling step is carried out according to the quantity of the degraded particles, so that the rationality of the distribution weight of the particles is completed. Finally, the simulation contrast process is added, the two-dimensional environment is accurately constructed, and the effectiveness and the superiority of the method are fully shown.
Description
Technical Field
The invention belongs to the technical field of synchronous positioning and mapping construction of mobile robots, and particularly relates to a mobile robot mapping method based on an improved Gmapping algorithm.
Background
In an unknown environment, the mobile robot body carrying the sensor such as the laser radar can acquire the experienced position, posture, map and the like through the laser radar, so the position and map at this time is fixed. The Gmapping algorithm is a particle filtering-based SLAM algorithm, and can effectively solve the problem of simultaneous positioning and building of a two-dimensional map, and the positioning and building of the map are separated. Each particle in Gmapping algorithm can carry a map, and by this method it is also more efficient to describe the actual position and pose of the mobile robot. The specific algorithm process is shown in fig. 1. The total number n of particles generated by the Gmapping algorithm initialization remains unchanged all the time during the process of building the two-dimensional map. Along with the continuous movement of the mobile robot, the map information carried by the particles also changes synchronously, but the map information of the positions of the mobile robot body is different, so that the data volume to be processed is different, and the number of the particles is relatively large in some simple positions, so that the map is accurately built, and excessive particles are caused, so that the computational resource is wasted. In contrast, in a complex location, the number of particles required is large, but the number of particles provided is relatively small, so that the mapping effect is poor and the environmental map cannot be fully represented.
Disclosure of Invention
The invention aims at optimizing an algorithm aiming at the situation of uneven distribution weights of particles, optimizing the algorithm to obtain an optimization algorithm for improving the distribution weights of the particles, and ensuring the uniformity of the particle weights.
In order to achieve the above purpose, the present invention adopts the following technical scheme.
A mobile robot mapping method based on an improved Gmapping algorithm comprises the following steps:
Step (1), acquiring data through an odometer and a laser radar, and judging two-dimensional environment information and position posture information of the position of the mobile robot body through the data after data processing, wherein the two-dimensional environment information and the position posture information are specifically:
Initializing to generate n initial particles, wherein each particle independently records a possible robot track and an environment map, and the states of the system are recorded by the particles; each particle has a weight of i＝1,2,…,n；
Step (2), updating and normalizing the weight of each particle according to the actual observed value at the moment k and the position and posture of each particle, and normalizing the weight of each particle
Step (3), setting a threshold N th according to the precision requirement, comparing the effective particle number N eff with a threshold N th, if N eff<Nth, indicating that the particles are degraded, carrying map information is missing, and the map cannot be fully represented, and returning to the step (2) for resampling operation; wherein the effective particle number A weight value representing the i-th particle at the k moment;
Step (4) adding noise to the moving position and posture information in the moving process of the robot, placing each particle around the mobile robot body, and carrying out scanning matching on each particle; for the environment information obtained from the laser radar combined with the pose of the mobile robot, updating the map information of the particles through the motion trail of the particles, judging whether the map information carried by the particles in the state is similar to a predicted value, if so, matching the map information with the position coordinates of each frame at the moment k one by one, and after determining the position coordinates of the matched particles, distributing the particles carrying the map around the current position coordinates; and positioning the degraded particles which do not contain map information, deleting the degraded particles, and updating the weight of the particles.
And (5) resampling, distributing particle weights, discarding particles with small weights, carrying out normalization weights on the particles, judging the particle states, updating the particle weights and map information thereof based on resampling results, and finishing map updating.
In the step (1), if the position and posture information of the mobile robot are not clear, the posture information of N particles is randomly generated and the postures of the robot are randomly arranged; if the initial position is already determined by the information of the particles, N particles are randomly generated and placed around the initial position, and the N initial particles are weightedThe particle characterization system records a possible robot track and an environment map;
The beneficial effects are that:
compared with the prior art, the method has better mapping effect on the 2D environment. Thanks to the key link for judging the state of the particles in the invention, if the degradation of the particles is serious, the resampling process of the particles is needed, so that all the particles can play a role. In a simple two-dimensional environment, the mobile robot can build a map with few particles, whereas in a more complex location it can describe the map with a large number of particles. By this means, the mobile robot can build a two-dimensional map of the location. The invention has accurate construction of the two-dimensional environment, can output the position map matched with the two-dimensional simulation environment, has less map noise and small map distortion.
Drawings
FIG. 1 is a basic flow chart of a mobile robot mapping algorithm;
FIG. 2 is a flow chart of a mobile robot mapping improvement algorithm;
FIG. 3 is a simulation diagram of the evaluation result;
FIG. 4 is a simulation diagram before an unmodified algorithm is based;
Fig. 5 is a simulation diagram based on the improved algorithm.
Detailed Description
Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.
The terms "comprising" and "having" and any variations thereof in the description embodiments of the invention and in the claims and drawings are intended to cover a non-exclusive inclusion, such as a series of steps or elements.
The technical scheme of the invention is further described in detail below with reference to the accompanying drawings and the examples.
As shown in fig. 1, the Gmapping algorithm process includes the following steps:
(1) And (5) data processing. After data processing is carried out by collecting data through an odometer and a laser radar, the mobile robot body judges two-dimensional environment information and position and posture information of the position of the mobile robot body through the data, n initial particles are generated in an initialized mode, the states of the system are recorded by the particles, and each particle independently records a possible robot track and an environment map. Wherein each particle has a weight of i＝1,2,…,n。
(2) And (5) predicting. According to the state and pose information x k-1 of the mobile robot body at the moment k-1, calculating a distribution probability density function p (x k|z1:k) at the moment k through Bayesian estimation, and accordingly predicting the moment k to obtain the state and pose information at the moment k. The Bayes theorem predicts that the equation is
(3) And (5) particle matching. And at the moment k, the mobile robot moves to a certain environment, whether map information carried by the particles in the state is similar to a predicted value is judged, if so, the map information is matched with the position coordinates of each frame at the moment k one by one, and the next step is continued.
(4) Particle distribution. And (3) determining the position coordinates of the matched particles, so that the particles carrying the map are distributed around the current position coordinates.
(5) The particle weights are updated. Since some particles do not show map information due to the existence of particle degradation, the weight of the particles is updated.
(6) And (5) updating particle resampling. By setting a threshold value N th, the effective particle number N eff is compared with the threshold value, if N eff<Nth, the particles are degraded, the carried map information is missing, and the map cannot be fully represented, so that the resampling operation is performed, namely, the step (2) is returned.
Effective particle countWherein/>The weight value of the i-th particle at the k time is shown.
(7) And updating the map. After resampling, the weight of the particles changes, the map information carried by the particles is updated, and the map is updated according to the two points.
As shown in fig. 2, the improved Gmapping algorithm process includes the steps of:
(1) Initializing. The mobile robot body is not clear of the position and the gesture information of the mobile robot body, and the gesture of the robot can be randomly arranged by randomly generating the gesture information of N particles, so that the weight of each initial particle is 1/N, and N is usually the multiple of 100 as the magnitude. If the mobile robot body has defined the initial position by the information of the particles, the particles are placed around the initial position, and the weights of the N initial particles are all
(2) And (5) predicting. And (3) obtaining a system model describing the movement of the robot in the step (1), adding noise to the position and posture information of the movement in the movement process of the mobile robot, placing each particle around the mobile robot body, and carrying out scanning matching on each particle.
(3) And (5) adjusting. And for the environment information obtained from the laser radar and the pose of the mobile robot, updating the map information of the particles through the motion trail of the particles, so that whether to update the weight value of each particle can be judged.
(4) Resampling. The resampling process is a process of reasonably distributing particle weights, particles with small weights are discarded due to the influence of particle degradation, and after the particles are subjected to normalized weights, the state of the particles is judged, and in the process, the total number N of the particles is kept unchanged.
The technical effects of the invention are further described by combining simulation experiments:
1. simulation conditions: all simulation experiments are realized in an ROS (robot operating system) with an operating system of Ubuntu16.04.5 and a hardware environment of GPUNvidiaGeForceGTX1080 Ti;
2. Simulation content and result analysis:
the simulation experiment of the invention is carried out with the existing map construction effect based on Gmapping algorithm, as shown in figure 3, and the experimental results are intuitively compared.
Compared with the prior art, the method has better mapping effect on the 2D environment. Thanks to the key link for judging the state of the particles in the invention, if the degradation of the particles is serious, the resampling process of the particles is needed, so that all the particles can play a role. In a simple two-dimensional environment, the mobile robot can build a map with few particles, whereas in a more complex location it can describe the map with a large number of particles. By this means, the mobile robot can build a two-dimensional map of the location.
The simulation comparison is carried out on the original algorithm and the improved algorithm, a simulation environment with the length, the width and the height of 30 meters, 10 meters and 2.5 meters is built in an ROS system by utilizing Gazebo, a blue three-dimensional cylinder represents a mobile robot, a black dot above blue is a laser radar, three black small cabinets are arranged in a fence, and the small cabinets can block the movement of the mobile robot. The simulation process is shown in fig. 3, and the simulation results are shown in fig. 4 and 5.
In the mobile robot map building simulation process, the mobile robot is controlled by a keyboard to perform a circle of motion around a simulation environment, the built map can display a two-dimensional environment of the position of the mobile robot, and finally the built map is output. And comparing simulation analysis of the two algorithms before and after improvement.
From fig. 4 and 5, it can be seen that the method is accurate in mapping the two-dimensional environment, can output a position map matched with the two-dimensional simulation environment, has less map noise and less map distortion, and fully shows the effectiveness and superiority of the method.
The embodiments of the present invention have been described in detail. However, the present invention is not limited to the above-described embodiments, and various modifications may be made within the knowledge of those skilled in the art without departing from the spirit of the present invention.
The foregoing detailed description of the invention has been presented for purposes of illustration and description, and it should be understood that the invention is not limited to the particular embodiments disclosed, but is intended to cover all modifications, equivalents, alternatives, and improvements within the spirit and principles of the invention.
Claims (2)
1. The mobile robot mapping method based on the improved Gmapping algorithm is characterized by comprising the following steps of:
Step (1), acquiring data through an odometer and a laser radar, and after data processing, judging two-dimensional environment information and position posture information of the position of the mobile robot body through the data, initializing to generate n initial particles, wherein the particles record the state of a system, and each particle independently records a possible robot track and an environment map; each particle has a weight of
Step (2), updating and normalizing the weight of each particle according to the actual observed value at the moment k and the position and posture of each particle, and normalizing the weight of each particle
Step (3), setting a threshold N th according to the precision requirement, comparing the effective particle number N eff with a threshold N th, if N eff<Nth, indicating that the particles are degraded, carrying map information is missing, and the map cannot be fully represented, and returning to the step (2) to perform weight updating operation; wherein the effective particle number A weight value representing the i-th particle at the k moment;
Step (4) adding noise to the moving position and posture information in the moving process of the robot, placing each particle around the mobile robot body, and carrying out scanning matching on each particle; for the environment information obtained from the laser radar combined with the pose of the mobile robot, updating the map information of the particles through the motion trail of the particles, judging whether the map information carried by the particles in the state is similar to a predicted value, if so, matching the map information with the position coordinates of each frame at the moment k one by one, and after determining the position coordinates of the matched particles, distributing the particles carrying the map around the current position coordinates; positioning degradation particles which do not contain map information, deleting the degradation particles, and updating the weight of the particles;
And (5) resampling, distributing particle weights, discarding particles with small weights, carrying out normalization weights on the particles, judging the particle states, updating the particle weights and map information thereof based on resampling results, and finishing map updating.
2. The mobile robot mapping method based on the improved Gmapping algorithm according to claim 1, wherein in the step (1), if the position and posture information of the mobile robot are not clear, the posture information of N particles is randomly generated to randomly arrange the postures of the robot; if the initial position is already determined by the information of the particles, N particles are randomly generated and placed around the initial position, and the N initial particles are weightedThe particles characterize the state of the system, recording a possible robot trajectory and environment map.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310203297.0A CN116309907B (en) | 2023-03-06 | 2023-03-06 | Mobile robot mapping method based on improved Gmapping algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310203297.0A CN116309907B (en) | 2023-03-06 | 2023-03-06 | Mobile robot mapping method based on improved Gmapping algorithm |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116309907A CN116309907A (en) | 2023-06-23 |
| CN116309907B true CN116309907B (en) | 2024-06-04 |
Family
ID=86799003
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310203297.0A Active CN116309907B (en) | 2023-03-06 | 2023-03-06 | Mobile robot mapping method based on improved Gmapping algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116309907B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111427370A (en) * | 2020-06-09 | 2020-07-17 | 北京建筑大学 | Sparse pose adjustment-based Gmapping mapping method for mobile robot |
| CN112785643A (en) * | 2021-02-02 | 2021-05-11 | 武汉科技大学 | Indoor wall corner two-dimensional semantic map construction method based on robot platform |
| CN112857379A (en) * | 2021-01-22 | 2021-05-28 | 南京邮电大学 | Improved Gmapping-SLAM map updating method and system |
| CN112882056A (en) * | 2021-01-15 | 2021-06-01 | 西安理工大学 | Mobile robot synchronous positioning and map construction method based on laser radar |
| CN113483747A (en) * | 2021-06-25 | 2021-10-08 | 武汉科技大学 | Improved AMCL (advanced metering library) positioning method based on semantic map with corner information and robot |
| CN114608585A (en) * | 2022-04-01 | 2022-06-10 | 常州大学 | Method and device for synchronous positioning and mapping of mobile robot |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9677890B2 (en) * | 2013-01-10 | 2017-06-13 | Intel Corporation | Positioning and mapping based on virtual landmarks |
-
2023
- 2023-03-06 CN CN202310203297.0A patent/CN116309907B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111427370A (en) * | 2020-06-09 | 2020-07-17 | 北京建筑大学 | Sparse pose adjustment-based Gmapping mapping method for mobile robot |
| CN112882056A (en) * | 2021-01-15 | 2021-06-01 | 西安理工大学 | Mobile robot synchronous positioning and map construction method based on laser radar |
| CN112857379A (en) * | 2021-01-22 | 2021-05-28 | 南京邮电大学 | Improved Gmapping-SLAM map updating method and system |
| CN112785643A (en) * | 2021-02-02 | 2021-05-11 | 武汉科技大学 | Indoor wall corner two-dimensional semantic map construction method based on robot platform |
| CN113483747A (en) * | 2021-06-25 | 2021-10-08 | 武汉科技大学 | Improved AMCL (advanced metering library) positioning method based on semantic map with corner information and robot |
| CN114608585A (en) * | 2022-04-01 | 2022-06-10 | 常州大学 | Method and device for synchronous positioning and mapping of mobile robot |
Non-Patent Citations (3)
| Title |
|---|
| A Review of Point Cloud Registration Methods Based on Laser SLAM;Jianggui Han等;《IEEE》;20230226;全文 * |
| 基于优化RBPF的同时定位与地图构建;罗元;苏琴;张毅;郑潇峰;;华中科技大学学报(自然科学版);20160523(第05期);全文 * |
| 基于自适应变换蝙蝠算法的机械臂轨迹优化;孙颙琰等;《机械传动》;20221231;第46卷(第5期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116309907A (en) | 2023-06-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109798896B (en) | Indoor robot positioning and mapping method and device | |
| CN110689576B (en) | Automatic ware-based dynamic 3D point cloud normal distribution AGV positioning method | |
| CN101276468B (en) | Computer implemented method for tracking object in sequence of frames of video | |
| Kersting et al. | Most likely heteroscedastic Gaussian process regression | |
| JP5183665B2 (en) | A decision mechanism, method, module and robot configured to determine at least one future activity of the robot | |
| Vasquez-Gomez et al. | View/state planning for three-dimensional object reconstruction under uncertainty | |
| US7929730B2 (en) | Method and system for object detection and tracking | |
| CN110333720A (en) | A kind of SLAM optimization method based on particle filter | |
| CN105182311B (en) | Omnidirectional's radar data processing method and system | |
| US20160195401A1 (en) | Method and system for locating an object | |
| Stachniss et al. | Mapping and exploration with mobile robots using coverage maps | |
| CN112882056A (en) | Mobile robot synchronous positioning and map construction method based on laser radar | |
| CN113108773A (en) | Grid map construction method integrating laser and visual sensor | |
| Senanayake et al. | Spatio-temporal hilbert maps for continuous occupancy representation in dynamic environments | |
| CN103901891A (en) | Dynamic particle tree SLAM algorithm based on hierarchical structure | |
| CN116088503B (en) | Dynamic obstacle detection method and robot | |
| CN111739066B (en) | Visual positioning method, system and storage medium based on Gaussian process | |
| Leung et al. | An improved weighting strategy for rao-blackwellized probability hypothesis density simultaneous localization and mapping | |
| CN116628448A (en) | Sensor management method based on deep reinforcement learning in extended target | |
| CN110702093B (en) | Particle filter-based positioning method and device, storage medium and robot | |
| Wang et al. | Virtual maps for autonomous exploration with pose SLAM | |
| CN114608585A (en) | Method and device for synchronous positioning and mapping of mobile robot | |
| CN116309907B (en) | Mobile robot mapping method based on improved Gmapping algorithm | |
| CN109121081B (en) | Indoor positioning method based on position candidate set and EM algorithm | |
| CN111274529B (en) | Robust Gao Sini Weisal PHD multi-expansion target tracking algorithm |
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 |