Communication system of underground space unmanned equipment
By designing an underground space unmanned equipment communication system that includes environment perception, decision control and communication scheduling subsystem, the problem of limited communication bandwidth of unmanned equipment in underground space is solved, high-speed data interaction and communication coverage are achieved, and the task completion ability of unmanned equipment in complex environments is improved.
Patent Information
- Application Number
- CN202411826265.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-12
- Publication Date
- 2025-03-28
AI Technical Summary
In underground space, communication and communication between unmanned devices and ground control centers is difficult, resulting in limited communication bandwidth and cannot support applications such as collaborative perception, distributed planning decision-making, and cloud-edge collaborative computing in complex environments, limiting the ability of unmanned intelligent devices to complete tasks in complex environments.
A communication system for unmanned equipment in underground space is designed, including communication unmanned vehicles, environmental perception subsystem, decision-making control subsystem and communication scheduling subsystem. Through the environment perception subsystem, a three-dimensional semantic map is constructed, and the decision-control subsystem adjusts communication strategies based on the map, optimizes the deployment location and motion trajectory of the unmanned vehicle to ensure communication coverage.
It realizes high-speed data interaction between unmanned intelligent devices in underground space, improves the ability of unmanned devices to complete tasks in complex environments, and ensures the reliability of communication coverage and information interaction.
Smart Images
Figure CN119729515A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of signal processing, and in particular to a communication system for unmanned equipment in an underground space. Background Art
[0002] With the development of robots and autonomous intelligent technology, robots and other unmanned equipment are widely used in human production and life. Especially in some special environments, such as underground spaces such as mines and tunnels, the environment is extremely harsh and threatens the life safety of operators. Robots and other unmanned intelligent equipment can replace humans to complete dangerous and heavy work in such environments, reducing personnel safety risks. However, underground space is usually not covered by public communication networks, and the environment structure is complex, making it difficult for unmanned equipment to communicate with the ground control center. Due to the difficulty and high cost of deploying fixed communication network equipment in underground space, the communication bandwidth of unmanned intelligent equipment such as underground robots is limited, which seriously restricts the information interaction ability between unmanned equipment and ground control consoles and between unmanned equipment. It is impossible to support the collaborative perception of complex environments in underground spaces, distributed planning and decision-making, cloud-edge collaborative computing and other applications that require high communication bandwidth, limiting the ability of unmanned intelligent equipment to complete tasks in complex environments. With the development of artificial intelligence technology, the ability of unmanned intelligent equipment in underground spaces to autonomously complete complex tasks has gradually increased, and communication capabilities have become the main bottleneck for the use of unmanned intelligent equipment. Underground wireless communication channels are affected by complex environmental structures and it is difficult to establish high-speed transmission channels. Task collaboration between multiple unmanned intelligent devices, sensor information entering the cloud, distributed decision-making, and human-machine shared control are difficult to achieve.
[0003] How to build a wireless communication network construction method in special environments such as underground space, adapt to the impact of complex environments on wireless communication systems, provide reliable information communication means for unmanned intelligent devices, and thus improve the ability of unmanned devices to complete tasks in complex spaces is a problem we need to solve. To this end, a method and system for building a communication network for unmanned devices in underground space is now provided. Summary of the invention
[0004] To achieve the above objectives, the present invention is implemented through the following technical solutions: A communication system for unmanned equipment in underground space, characterized by: comprising a communication unmanned vehicle, an environment perception subsystem, a decision control subsystem and a communication scheduling subsystem; The environmental perception subsystem includes an environmental information acquisition module, an environmental perception preprocessing module, and an environmental modeling module; The communication unmanned vehicle includes an unmanned vehicle platform, vehicle-mounted sensors, an unmanned vehicle-mounted base station and an unmanned vehicle base station inter-station interconnection device; The environmental information acquisition module is used to collect and aggregate the perception information of the surrounding environment acquired by the communication unmanned vehicle and other unmanned intelligent equipment; The environmental information preprocessing module is used to preprocess the collected environmental data to obtain an image signal after eliminating the influence of noise and distortion; The environment modeling module is used to identify and understand the image signal obtained after eliminating noise and distortion to obtain an environment structure model; The communication decision subsystem is used to adjust the communication strategy according to the environment model; The planning and control subsystem is used to generate decision instructions to control unmanned equipment.
[0005] Preferably, the process of collecting environmental signals by the environmental perception module includes: The environment perception module is provided with an image signal acquisition terminal, which is installed at a corresponding position of the unmanned equipment. When an environment perception instruction is received, the environment characteristics are acquired through the installed image signal acquisition terminal.
[0006] Preferably, the process of preprocessing the collected environmental image signal by the environmental perception signal preprocessing module includes: Acquire the noise signal contained in the collected environmental image data, classify it into environmental noise data, acquire the collected environmental image signal, and generate environmental structure data; The environmental image data of the acquired environmental structure data are fused and spliced by multi-source fusion method, and the local environmental map collected according to the time series is fused and disambiguated to generate a large-scale environmental map; The errors caused in the map construction process are eliminated through loop closure detection, and a large-scale environment map is obtained after the errors in the map construction process are eliminated.
[0007] Preferably, the process of the environment modeling module modeling and identifying the obtained large-scale environment map includes: The foreground objects and background objects in the collected image information are segmented through the semantic segmentation model to obtain an image with object classification annotations; Associating the obtained image containing the object annotation classification with the image obtained at the adjacent time, and estimating the pose estimation value of the object annotation image in the three-dimensional space; Cross-validate the position and attitude estimation of the obtained image containing object classification annotations in three-dimensional space with the attitude information obtained by the inertial sensor to reduce the error range of visual attitude estimation; According to the position and posture in the three-dimensional space, the obtained image containing the object classification is projected into a large-scale global map, and a semantic map with object classification annotations is constructed in the three-dimensional space.
[0008] Preferably, the process of adjusting the communication strategy according to the environment model by the communication decision subsystem includes: Based on the semantic map with object classification annotations in the three-dimensional space, the influence of the materials of different types of objects in the three-dimensional space on the attenuation of wireless signals is comprehensively considered to construct a wireless communication signal strength heat map. Based on the wireless signal strength heat map, the area where the wireless signal coverage strength is lower than a certain threshold is searched; According to the communication connection constraints between the communication unmanned vehicle base stations and based on the IAB signal strength measurement, the signal strength threshold constraint conditions between the communication unmanned vehicles and the distance constraint conditions between the unmanned vehicles are established; The wireless signal strength at the location is measured by unmanned intelligent devices, and reported to the communication dispatch subsystem, and the location of the intelligent devices in the environment space whose signal strength is lower than the set threshold is calculated; Based on the wireless signal strength heat map and the signal strength received by the smart device, the signal coverage quality measurement in the space is calculated, and the deployment position of the communication unmanned vehicle is optimized through a dynamic optimization algorithm with the goal of minimizing the coverage blind area.
[0009] Preferably, the process of generating unmanned equipment control instructions by the planning control subsystem includes: According to the communication connection constraints between the communication unmanned vehicle base stations and based on the IAB signal strength measurement, the signal strength threshold constraint conditions between the communication unmanned vehicles and the distance constraint conditions between the unmanned vehicles are established; According to the semantic map with object classification annotations in three-dimensional space, the movement trajectory of the unmanned platform in the process of completing the task is planned according to the communication distance constraint, so that the unmanned equipment always meets the communication distance constraint in the process of completing the task. At the same time, according to the coverage range of the unmanned vehicle, by obtaining the position of the unmanned equipment in the environment space, the communication coverage of the unmanned equipment in the task space is guaranteed.
[0010] The present invention provides a communication system for unmanned equipment in underground space. It has the following beneficial effects: The communication system of the underground unmanned equipment performs perception modeling on the environment space where the unmanned equipment performs tasks, pre-processes the collected images to obtain a local map that eliminates environmental noise, and fuses the local maps to obtain a global map of a large-scale scene. The foreground objects and background objects in the collected image information are segmented by a semantic segmentation model to obtain an image with object classification annotations; the visual pose estimation is performed on the obtained image with object annotation classification to estimate the pose estimation value of the object annotation image in three-dimensional space; the position and pose estimation of the obtained image in three-dimensional space are cross-validated with the pose information obtained by the inertial sensor to reduce the error range; the obtained image containing object classification is projected onto a large-scale global map to construct a semantic map with object classification annotations in three-dimensional space. The planning and decision-making subsystem plans the deployment location and motion trajectory of the communication unmanned vehicle base station based on the generated three-dimensional semantic map, and generates the control instructions of the unmanned vehicle. A wireless signal strength heat map is established based on the received signal strength of the unmanned equipment, and a communication blind spot filling strategy is generated based on the heat map to adjust the base station transmission power and the deployment location of the unmanned vehicle base station.
[0011] The present invention proposes a method for a communication unmanned vehicle to provide efficient network coverage for unmanned equipment in underground space, so that the unmanned intelligent equipment can achieve high-speed data interaction, thereby improving the ability of the unmanned intelligent equipment to complete tasks in underground space. BRIEF DESCRIPTION OF THE DRAWINGS
[0012] Figure 1 It is a schematic diagram of the communication system of the unmanned equipment in the underground space of the present invention. DETAILED DESCRIPTION
[0013] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0014] See also Figure 1 , the present invention provides a technical solution: a communication system for unmanned equipment in underground space, including a communication unmanned vehicle, an environment perception subsystem, a decision control subsystem and a communication scheduling subsystem; the environment perception subsystem includes an environment information acquisition module, an environment perception preprocessing module and an environment modeling module; The communication unmanned vehicle includes an unmanned vehicle platform, vehicle-mounted sensors, an unmanned vehicle-mounted base station and an unmanned vehicle base station inter-station interconnection device; The environmental information acquisition module is used to collect and aggregate the perception information of the surrounding environment acquired by the communication unmanned vehicle or other unmanned intelligent equipment; It should be further explained that, in the specific implementation process, the process of the environmental information collection module collecting the surrounding environment information includes: The environment perception module is provided with an image signal acquisition terminal, which is installed at a corresponding position of the unmanned equipment. When an environment perception instruction is received, the environment characteristics are acquired through the installed image signal acquisition terminal.
[0015] The communication unmanned vehicle includes a vehicle-mounted sensor with a variety of built-in photoelectric sensing devices. After starting to work, it senses the surrounding three-dimensional environmental information, adds RGB images and depth images to timestamps, and compresses them; The environmental information preprocessing module is used to preprocess the collected environmental data to obtain an image signal after eliminating the influence of noise and distortion; It should be further explained that, in the specific implementation process, the process of preprocessing the environmental information by the environmental information preprocessing module includes: Acquire the noise signal contained in the collected environmental image data, classify it into environmental noise data, acquire the collected environmental image signal, and generate environmental structure data; The two-dimensional and three-dimensional image information of multiple communicating unmanned vehicle onboard sensors at the same time are fused, feature points are extracted from three-dimensional data of different perspectives, and multi-perspective alignment and splicing are performed.
[0016] The environmental image data of the acquired environmental structure data are fused and spliced by multi-source fusion method, and the local environmental map collected according to the time series is fused and disambiguated to generate a large-scale environmental map; The errors caused in the map construction process are eliminated through loop closure detection, and a large-scale environment map is obtained after the errors in the map construction process are eliminated.
[0017] The environment modeling module is used to identify and understand the image signal obtained after eliminating noise and distortion to obtain an environment structure model; It should be further explained that, in the specific implementation process, the process of the environment modeling module identifying and understanding the image signal to obtain the environment structure model includes: The foreground objects and background objects in the collected image information are segmented through the semantic segmentation model to obtain an image with object classification annotations; Associating the obtained image containing the object annotation classification with the image obtained at the adjacent time, and estimating the pose estimation value of the object annotation image in the three-dimensional space; Cross-validate the position and attitude estimation of the obtained image containing object classification annotations in three-dimensional space with the attitude information obtained by the inertial sensor to reduce the error range of visual attitude estimation; According to the position and posture in the three-dimensional space, the obtained image containing the object classification is projected into a large-scale global map, and a semantic map with object classification annotations is constructed in the three-dimensional space.
[0018] Furthermore, the process of generating unmanned equipment control instructions by the planning control subsystem includes: In the 3D semantic map with object classification annotations, the optimal deployment range of the communication unmanned vehicle is planned according to the coverage distance of the unmanned vehicle base station and the communication distance between the interconnected devices of the unmanned vehicle base station. With the goal of maximizing the communication coverage range, the optimal deployment position of the unmanned vehicle is planned taking into account the obstacle occlusion factor.
[0019] Based on the semantic map with object classification annotations in three-dimensional space and the communication distance constraints of interconnected devices between base stations, the deployment position of the communication unmanned vehicle in the mission space is planned, so that the communication unmanned vehicle always meets the communication distance constraints when completing the mission.
[0020] The Harris corner detection algorithm is used to detect the corner positions in the task environment map and identify the positions that are likely to cause wireless signal interruption. These positions are used as the n pre-selected positions for the deployment of the communication unmanned vehicle, denoted as M={m1,m2,…mn}.
[0021] Based on the object category segmentation results in the semantic map, the pre-selected positions M of the communication unmanned vehicle are divided into two categories through semantic map labels. One type of pre-selected positions is surrounded by objects that cannot be penetrated by signals, and the set of corresponding positions is recorded as M1. The other type of pre-selected positions is composed of objects that can be penetrated by wireless signals, and the set of corresponding positions is recorded as M2.
[0022] The points in the set M1 are used as the locations where the communicating unmanned vehicles are deployed. The distance Dij between the communicating unmanned vehicles is calculated and traversed. It is assumed that the maximum communication distance between the interconnected devices of the unmanned vehicle-mounted base stations is Dmax. When the distance Dij between the two communicating unmanned vehicles is greater than Dmax, an additional communicating unmanned vehicle is added between the two corresponding communicating unmanned vehicles as a relay.
[0023] At the same time, according to the coverage of the unmanned vehicle, the communication coverage of the unmanned equipment in the mission space is guaranteed by obtaining the position of the unmanned equipment in the environmental space. The intelligent unmanned equipment in the environmental space obtains its own position through the constructed three-dimensional semantic map and reports it to the planning and control subsystem. The planning and control subsystem calculates the distribution position of the unmanned intelligent equipment through the clustering algorithm to achieve the optimal deployment position.
[0024] The communication decision subsystem is used to adjust the communication strategy according to the environment model.
[0025] Furthermore, the process of the communication decision subsystem adjusting the communication strategy according to the environment model includes: According to the semantic map with object classification annotations in three-dimensional space, the influence of the materials of different types of objects in three-dimensional space on wireless signal attenuation, as well as the distribution position and transmission power of communication drones, is constructed to construct the wireless communication signal strength heat map G. Let q be the position in three-dimensional space, and the wireless signal strength map can be expressed as: G(q)=G1(q)+G2(q)+G3(q)+....+Gn(q) Among them, G1(q), G2(q), G3(q), ...Gn(q) are the distribution functions of n communicating unmanned vehicles in the environment space.
[0026]
[0027]
[0028] Where PT is the base station transmission power, GT is the transmission gain of the base station antenna carried by the communication unmanned vehicle, GR is the gain of the receiving antenna of the intelligent unmanned device as the terminal, λ is the wavelength of the wireless signal carrier, and d is the straight-line distance between the communication unmanned vehicle and the intelligent unmanned device as the terminal. The position of the i-th communication unmanned vehicle is recorded as
[0029] The straight-line distance from any point in the map to the communication unmanned vehicle base station i is:
[0030] Based on the wireless signal strength heat map, search for areas where the wireless signal coverage strength is below a certain threshold.
[0031] Among them, the wireless signal strength heat map, through the three-dimensional semantic map W, the position Q of the unmanned intelligent equipment in the map, the communication unmanned vehicle base station transmission power P and the signal attenuation factor.
[0032]
[0033]
[0034] Where η is the signal strength attenuation coefficient, and the signal fading model is used for calculation. d is the length of the straight-line transmission path between the terminal and the base station, that is, the straight-line distance between the terminal and the base station, db is the length of the obstacle in the straight-line transmission path, ɑ is a coefficient related to the obstacle signal absorption performance, and its value is [0,1]. When the obstacle completely blocks or absorbs the wireless signal, ɑ=1; when the obstacle does not block or absorb the wireless signal, ɑ=0.
[0035] The wireless signal strength at the location is measured by unmanned intelligent devices, and reported to the communication dispatch subsystem, and the location of the intelligent devices in the environment space whose signal strength is lower than the set threshold is calculated; Based on the wireless signal strength heat map and the signal strength received by the smart device, the area with the signal coverage quality area setting threshold in the space is jointly calculated, and the deployment position of the communication unmanned vehicle is optimized with the goal of minimizing the coverage blind area through a dynamic optimization algorithm.
[0036] The planning and control subsystem is used to generate decision instructions to control unmanned equipment.
[0037] According to the communication connection constraints between the communication unmanned vehicle base stations, the signal strength threshold constraint conditions between the communication unmanned vehicles and the distance constraint conditions between the unmanned vehicles are established.
[0038] According to the semantic map with object classification annotations in three-dimensional space, the movement trajectory of the unmanned platform in the process of completing the task is planned according to the communication distance constraint, so that the unmanned equipment always meets the communication distance constraint in the process of completing the task. At the same time, according to the coverage range of the unmanned vehicle, by obtaining the position of the unmanned equipment in the environment space, the communication coverage of the unmanned equipment in the task space is guaranteed.
[0039] Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field and related fields without creative work should fall within the scope of protection of the present invention. The structures, devices and operating methods not specifically described and explained in the present invention are implemented according to the conventional means in the field unless otherwise specified and limited.
Claims
1. A communication system for unmanned equipment in underground space, characterized in that: Including communication unmanned vehicle, environment perception subsystem, decision control subsystem and communication dispatch subsystem; The environmental perception subsystem includes an environmental information acquisition module, an environmental perception preprocessing module, and an environmental modeling module; The communication unmanned vehicle includes an unmanned vehicle platform, vehicle-mounted sensors, an unmanned vehicle-mounted base station and an unmanned vehicle base station inter-station interconnection device; The environmental information acquisition module is used to collect and aggregate the perception information of the surrounding environment acquired by the communication unmanned vehicle and other unmanned intelligent equipment; The environmental information preprocessing module is used to preprocess the collected environmental data to obtain an image signal after eliminating the influence of noise and distortion; The environment modeling module is used to identify and understand the image signal obtained after eliminating noise and distortion to obtain an environment structure model; The communication decision subsystem is used to adjust the communication strategy according to the environment model; The planning and control subsystem is used to generate decision instructions to control unmanned equipment.
2. A communication system for unmanned equipment in underground space according to claim 1, characterized in that: The process of collecting environmental signals by the environmental perception module includes: The environment perception module is provided with an image signal acquisition terminal, which is installed at a corresponding position of the unmanned equipment. When an environment perception instruction is received, the environment characteristics are acquired through the installed image signal acquisition terminal.
3. A communication system for unmanned underground space equipment according to claim 2, characterized in that: The process of preprocessing the collected environmental image signal by the environmental perception signal preprocessing module includes: Acquire the noise signal contained in the collected environmental image data, classify it into environmental noise data, acquire the collected environmental image signal, and generate environmental structure data; The environmental image data of the acquired environmental structure data are fused and spliced by multi-source fusion method, and the local environmental map collected according to the time series is fused and disambiguated to generate a large-scale environmental map; The errors caused in the map construction process are eliminated through loop closure detection, and a large-scale environment map is obtained after the errors in the map construction process are eliminated.
4. A communication system for unmanned equipment in underground space according to claim 3, characterized in that: The process of modeling and identifying the obtained large-scale environmental map by the environmental modeling module includes: The foreground objects and background objects in the collected image information are segmented through the semantic segmentation model to obtain an image with object classification annotations; Associating the obtained image containing the object annotation classification with the image obtained at the adjacent time, and estimating the pose estimation value of the object annotation image in the three-dimensional space; Cross-validate the position and attitude estimation of the obtained image containing object classification annotations in three-dimensional space with the attitude information obtained by the inertial sensor to reduce the error range of visual attitude estimation; According to the position and posture in the three-dimensional space, the obtained image containing the object classification is projected into a large-scale global map, and a semantic map with object classification annotations is constructed in the three-dimensional space.
5. The communication system for unmanned equipment in underground space according to claim 4, characterized in that: The process of the communication decision subsystem adjusting the communication strategy according to the environment model includes: Based on the semantic map with object classification annotations in the three-dimensional space, the influence of the materials of different types of objects in the three-dimensional space on the attenuation of wireless signals is comprehensively considered to construct a wireless communication signal strength heat map. Based on the wireless signal strength heat map, the area where the wireless signal coverage strength is lower than a certain threshold is searched; According to the communication connection constraints between the communication unmanned vehicle base stations and based on the IAB signal strength measurement, the signal strength threshold constraint conditions between the communication unmanned vehicles and the distance constraint conditions between the unmanned vehicles are established; The wireless signal strength at the location is measured by unmanned intelligent devices, and reported to the communication dispatch subsystem, and the location of the intelligent devices in the environment space whose signal strength is lower than the set threshold is calculated; Based on the wireless signal strength heat map and the signal strength received by the smart device, the signal coverage quality measurement in the space is calculated, and the deployment position of the communication unmanned vehicle is optimized through a dynamic optimization algorithm with the goal of minimizing the coverage blind area.
6. The communication system for unmanned equipment in underground space according to claim 5, characterized in that: The process of generating unmanned equipment control instructions by the planning control subsystem includes: According to the communication connection constraints between the communication unmanned vehicle base stations and based on the IAB signal strength measurement, the signal strength threshold constraint conditions between the communication unmanned vehicles and the distance constraint conditions between the unmanned vehicles are established; Based on the semantic map with object classification annotations in three-dimensional space, the movement trajectory of the unmanned platform in the process of completing the task is planned according to the communication distance constraint, so that the unmanned equipment always meets the communication distance constraint in the process of completing the task. At the same time, according to the coverage range of the unmanned vehicle, the position of the unmanned equipment in the environmental space is obtained to ensure the communication coverage of the unmanned equipment in the task space.