WO2020006698A1 - 一种土石方智能碾压系统 - Google Patents

一种土石方智能碾压系统 Download PDF

Info

Publication number
WO2020006698A1
WO2020006698A1 PCT/CN2018/094460 CN2018094460W WO2020006698A1 WO 2020006698 A1 WO2020006698 A1 WO 2020006698A1 CN 2018094460 W CN2018094460 W CN 2018094460W WO 2020006698 A1 WO2020006698 A1 WO 2020006698A1
Authority
WO
WIPO (PCT)
Prior art keywords
compaction
rolling
roller
unmanned
quality
Prior art date
Application number
PCT/CN2018/094460
Other languages
English (en)
French (fr)
Inventor
李庆斌
张庆龙
安再展
刘天云
Original Assignee
清华大学
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 清华大学 filed Critical 清华大学
Publication of WO2020006698A1 publication Critical patent/WO2020006698A1/zh

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
    • G05D1/0257—Control of position or course in two dimensions specially adapted to land vehicles using a radar
    • 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
    • G05D1/0212—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory
    • G05D1/0223—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory involving speed control of the vehicle
    • 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
    • G05D1/0276—Control of position or course in two dimensions specially adapted to land vehicles using signals provided by a source external to the vehicle
    • 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
    • G05D1/0276—Control of position or course in two dimensions specially adapted to land vehicles using signals provided by a source external to the vehicle
    • G05D1/0278—Control of position or course in two dimensions specially adapted to land vehicles using signals provided by a source external to the vehicle using satellite positioning signals, e.g. GPS

Definitions

  • the invention belongs to the technical field of earth and stone engineering, and particularly relates to an intelligent earth and stone rolling system, which is mainly used for quality control of earth and stone engineering filling and rolling in dams, highways, railways, airports, dams, bridges, ports and other fields.
  • the stability of earthwork is determined by the quality of the filling construction.
  • the specific control indicators include result control indicators (such as compaction K, dry density ⁇ , etc.), source control indicators (such as water content w, graded p, etc.) and construction.
  • Parameter control indicators such as rolling number n, driving speed v, vibration frequency f, driving trajectory, etc.
  • Earthwork compaction monitoring system is an effective means to realize the whole process monitoring of the above three indicators.
  • the related system has been studied by scholars at home and abroad and some results have been achieved. Among them, Roller-Integrated Compaction Monitoring (RICM) has been widely used in earth and rock engineering in the United States and Europe.
  • RCM Roller-Integrated Compaction Monitoring
  • foreign compaction monitoring systems based on RICM technology mainly include the rolling process monitoring system based on the drive power (MDP) of the American Caterpillar company, the ACE (Ammann Compaction) based on the soil stiffness characterization index K s of the Swedish Ammann company. Expert) system, Dynamic and Trimble's RICM system based on the acceleration characterization index CMV, and Rinehart and Mooney's RICM system based on the total harmonic distortion index (THD).
  • Continuous Compaction Control is a data acquisition system installed on compaction equipment that can continuously collect real-time information on the operation and performance of the compactor. In a sense, CCC is another name for RICM.
  • AFC automatic feedback control system
  • the amplitude and frequency in the system can be automatically Adjustment, using the AFC system to compact the soil can increase the opportunity for rapid compaction and improve the uniformity of soil properties.
  • AFC automatic feedback control
  • IC Intelligent Compaction
  • FHWA Federal Highway Administration
  • IC refers to the use of modern vibratory rollers equipped with integrated measurement systems, on-board computer reporting systems, map-based global positioning systems (GPS), and feedback control systems for road materials.
  • Compaction systems such as soil, aggregate or asphalt pavement materials.
  • IC technology has been implemented in Europe and Japan for many years, and the technology was introduced to the United States in late 2000. Dynamic control of machine parameters can apply vibration energy to under-pressure areas, prevent over-compaction and ensure uniform compaction of soil / aggregate. Although the application of IC technology and its systems is promising, their performance has not been fully evaluated.
  • Liu et al. Used the real-time monitoring system of the compaction quality of the earth-rock dam to collect the construction parameters (rolling number n, compaction thickness h, driving trajectory, driving speed v) in real time, and then analyzed the water content w, gradation p. Analysis of the correlation between compaction quality (such as compaction degree K, dry density ⁇ ) and roller compaction construction parameters.
  • the model combined with the Kriging space interpolation method, proposes a fast evaluation method for the full working face of the compaction quality of earth-rock dam filling materials based on the CV value, and achieves a fast evaluation of the full working face compaction quality.
  • the object of the present invention is to provide an intelligent earth-rock-rolling compaction system.
  • the system is based on the material-machine-information-machine autonomous decision interaction system framework, and applies intelligent decision control methods to compaction quality monitoring.
  • Real intelligent compaction is realized in the system.
  • the system can be used for quality control of earthwork construction and filling. As an effective solution for intelligent control of roller compaction quality during the construction of earthwork, it can have machine autonomy for roller compaction quality.
  • the system mainly includes an unmanned rolling system and an automatic decision-making control system for the working parameters of the roller, which realizes the automatic decision-making control of the compaction parameter machine to achieve intelligent and efficient rolling.
  • An earth-rock intelligent roll compacting system is implemented based on a material-machine-information-machine autonomous decision interaction system framework, including:
  • the real-time compaction condition detection data includes the compaction effect index value S (x, y) of the position and the working parameters of the roller
  • the working parameters of the roller include the vibration frequency f (x, y), and the amplitude A (x, y) and driving speed V (x, y), and may also include excitation force, etc .
  • Unmanned roller compactor and its cooperative rolling operation system in which the unmanned roller compactor performs unmanned roller compaction operations according to the rolling operation parameters;
  • RTK-system providing high-precision positioning and navigation services for unmanned rollers
  • the wireless communication system provides real-time high-speed communication services for a single unmanned roller compactor or a group of unmanned roller compactors' cooperative rolling operation systems, and provides real-time communication services for unmanned roller compactors and remote quality monitoring centers. In addition, It also provides services for other scenarios that require communication during the rolling construction operation;
  • the remote quality monitoring center plans the operation area and navigation route based on the three-dimensional digital model of the construction of earth and stone engineering. It receives the operation data and video information of the operation environment through the wireless communication system, and sends automatic information to the operation. Navigation data and instructions, man-machine interaction, remote operation of unmanned rollers according to receiving emergency processing requests, provision of scheduling services for unmanned rollers and clusters of coordinated rolling operation systems, and display of on-site rolling construction operations, real-time storage On-site rolling construction data.
  • the autonomous decision control system for working parameters of a compactor mainly includes an independent decision model training module for the working parameters of the compactor and a predicting module for the working parameters of the compactor.
  • the independent decision model training module for the working parameters of the compactor uses an artificial intelligence algorithm such as (Machine learning algorithm) model training data to generate a prediction model
  • the rolling machine working parameter prediction module uses the input parameters and the model training trained periodically to obtain the working parameters of the rolling machine with the maximum efficiency that will be required in the next local area, and formulate the compaction strategy To achieve intelligent control of the compaction process.
  • roller compactor working parameter independent decision model training module and roller compactor working parameter prediction module can also be called compactor intelligent compaction model module: the intelligent compaction model mainly establishes the relationship between compaction effect and compaction parameters, and according to the current Compaction conditions, optimization of rolling parameters, formulation of compaction strategies, and intelligent control of compaction processes.
  • the training module auto-decision model training module for working parameters uses data obtained by a real-time positioning device, an on-line compaction quality detection device, and a sensor device, including the position coordinates (x, y) and real-time compaction condition detection data to form training data
  • a real-time positioning device an on-line compaction quality detection device
  • a sensor device including the position coordinates (x, y) and real-time compaction condition detection data to form training data
  • the working parameters of the roller compactor, the vibration frequency f n-1 (x, y), the amplitude A n-1 (x, y), and the speed V n-1 (x, y) constitute the input data of the training model.
  • the data model is trained machine learning algorithm to obtain the relationship between compaction and grinding operating parameters, namely:
  • S n (x, y) F (S n-1 (x, y), f n-1 (x, y), A n-1 (x, y), V n-1 (x, y))
  • the roller compactor working parameter prediction module When the roller compactor working parameter prediction module operates in real time, it will obtain the current compaction status S n-1 (x, y) from the real-time positioning device, the compaction quality online detection device, and the sensor device in real time. ) To optimize the compaction parameters with the goal of achieving the highest efficiency of the specified compaction effect index S E (x, y) to obtain the output of the intelligent compaction model:
  • f n (x, y), A n (x, y), V n (x, y) are the vibration frequency of the roller, the amplitude of the roller, and the speed of the roller in the next rolling operation to achieve the rolling operation Parameter autonomous decision control.
  • the autonomous decision-making control system for the working parameters of the roller compactor is also used for the quality assessment, quality feedback control (QC), and quality acceptance (QA) of the roller compaction for earth-rock engineering works.
  • the compaction quality control standard, the obtained compaction effect index value is calculated through the compaction quality evaluation model to calculate the corresponding design compaction compaction quality control index value, and then calculate whether the value meets the design compaction compaction quality control
  • the standard can thus achieve the assessment of the compaction quality of earth-rock works.
  • the independent decision control system of the working parameters of the compactor can be used for feedback control and quality acceptance of the compaction quality.
  • the machine has independent decision-making for quality assessment and quality feedback control. (QC) and quality acceptance (QA) intelligent features.
  • the unmanned roller compactor has functions such as remote wake-up, hibernation, high-precision RTK-system positioning and navigation, safety obstacle avoidance, independent planning of roller compaction operations, single or multi-machine cooperative autonomous roller compaction, and automatic detection of construction quality.
  • the sensing system, control system, driving system and roller body, the sensing system, driving system and control system form an on-board automatic control system.
  • the on-board automatic control device receives real-time automatic navigation instructions and remote driving from the remote quality monitoring center. The information is compared with the speed, steering angle, position and other information detected by the sensing system and RTK-system. The deviation is adjusted by the PID algorithm and transmitted to the electric steering wheel, electric throttle, electric brake and other actuators to control the vibration roller. Complete the desired driverless rolling operation.
  • the unmanned roller compactor can be controlled by the unmanned roller control module:
  • This module is mainly based on the real-time positioning device, wireless communication system, intelligent computing device, and roller control device to work together.
  • the real-time positioning device uses a mobile positioning measurement tool to build a three-dimensional planar geographic coordinate system of the work area before the roller operation.
  • the intelligent computing device will build the rolling motion track of the unmanned roller in the work area based on this three-dimensional planar geographic coordinate system.
  • the intelligent path planning algorithm of the intelligent computing device will be used to send control instructions to the compactor control device through the data line to achieve unmanned functions, such as automatically controlling the ignition, steering, speed, and braking of the compactor. , U-turn, etc.
  • the RTK (Real-time kinematic) system is an RTK-GPS system or an RTK-BDS system. Based on carrier phase difference technology and GPS (or BDS) latitude and longitude data, cm-level high-precision positioning is achieved.
  • the RTK-system acquires High-precision global latitude and longitude data is used to construct a three-dimensional planar geographic coordinate system, which is then converted into a high-precision planar geographic coordinate system by a mapping algorithm. It is mainly composed of GPS or BDS receiver, RTK-GPS or BDS reference station, on-site rover station, satellite, etc., and provides accurate space position information, driving speed information and other operating status information of the unmanned roller in real time. Calculate the distance deviation and angle deviation between the vehicle body and the optimal path, so as to calculate the adjusted steering angle, and provide high-precision positioning and navigation services for unmanned rollers.
  • the sensing system includes an anti-collision radar device and a rolling wheel sensing device.
  • the control system includes an intelligent computing device, an on-line compaction quality detection device, and a rolling machine control device.
  • the anti-collision radar device is installed in the rolling machine. At the forefront, the spatial distance of obstacles in the direction of the roller is monitored in real time. When the obstacle distance from the direction of movement reaches the warning threshold, the unmanned roller will adopt an intelligent avoidance strategy.
  • the on-line compaction quality detection device is composed of a sensor, a sensor installation device, a data acquisition card, and a data processing and analysis device, and belongs to the on-board measurement technology of the compaction quality of the filling construction. By measuring the compaction with the filling material during the compaction process, Quality-related signals.
  • Real-time online access to the compaction effect indicators of the filling material namely RICMMV, ICMV or SCV
  • the compaction effect indicators to characterize the compaction quality of the filling material.
  • Different compaction effect indicators can be used to characterize the filling material.
  • Building material compaction quality (such as stiffness, strength, deformation, settlement, temperature, density, or compaction); the on-line compaction quality detection device periodically detects the current compaction effect index value at the current position and the current position The information is obtained through the positioning device on the compactor, and the compaction effect indicator value is transmitted to the remote quality monitoring center through a wireless device.
  • the compaction effect indicator value is displayed with a visual effect.
  • the wireless communication system is mainly composed of an on-board communication device, a communication transfer station, and a remote monitoring platform communication system, which realizes the transmission of information and instructions between the unmanned roller and the remote quality monitoring center, and realizes the on-board automatic control device. Information communication between various parts, multiple unmanned rollers, and other nearby construction equipment.
  • the unmanned roller compactor group cooperative rolling operation system is based on the unmanned roller compactor, the remote quality monitoring center multi-machine dispatching system, the construction site path planning algorithm, and the RTK-GPS or RTK-BDS navigation technology, and remote quality monitoring
  • the center develops navigation routes and provides them to the automatic control device of the unmanned roller through a wireless communication system.
  • the real-time working parameters of the roller obtained from the autonomous decision control system of the roller control system control the execution of the unmanned roller.
  • Multi-machine cooperative intelligent rolling operation and automatic rolling, the remote quality monitoring center multi-machine scheduling system gives an unmanned rolling machine scheduling implementation plan according to the actual situation on the site to achieve effective scheduling of multiple unmanned rolling machines.
  • the unmanned roller compactor group provides basic services in cooperation with the rolling operation; the engineering site path planning algorithm provides an optimal rolling operation and obstacle avoidance navigation path for the unmanned roller compactor.
  • the remote quality monitoring center is a big data digital visualization platform for the operation area and an interactive platform with unmanned rollers, which can real-time display the conditions of the rolling construction operation site, and perform man-machine interaction to remotely control the unmanned rollers in emergency or special situations.
  • Machine operation in which the situation of the compaction construction job site includes the compaction quality information of the on-site filling materials (such as the continuous characterization index information of compaction quality), and the compaction construction parameter information (such as driving speed, number of compaction passes, compaction thickness, vibration Frequency) and live environment video information.
  • the invention has the characteristics of intelligence, high efficiency, real-time, high control accuracy, etc., and can perform real-time autonomous decision-making control of the rolling parameters during the next rolling operation according to the current compaction status, thereby realizing the autonomous decision-making control of the machine during the rolling process.
  • It is suitable for the quality control of the filling construction of earth and stone engineering.
  • the continuous compaction index value that characterizes the compaction quality of the filling material can be quickly detected to determine the compaction quality of the filling material and calculate the compaction model parameters.
  • the working parameters of the roller are optimized, and the optimal solution for compaction decision is proposed to control the working parameters of the roller.
  • the invention can be applied to the quality control of filling and construction of earth and rock works such as water conservancy, highways, railways, airports, ports, etc., not only can effectively control the quality of roller compaction, but also effectively improve the efficiency of roller compaction, and realize intelligent compaction control of the roller compaction , To achieve fine management of earthwork construction
  • FIG. 1 is a framework diagram of an intelligent compaction decision system provided by the present invention.
  • FIG. 2 is a schematic diagram of the intelligent compaction decision system provided by the present invention.
  • FIG. 3 is a software system flowchart of the intelligent compaction decision-making system provided by the present invention.
  • FIG. 4 is a structural diagram of an embodiment of an unmanned roller compactor provided by the present invention.
  • the invention provides an intelligent earth-rock-rolling compaction system, which can be used for quality control of earth-rock engineering filling construction.
  • the system is based on a material-machine-information-machine autonomous decision-making interactive system framework, and is used as a rock-roller during earth-rock engineering construction.
  • An effective solution for intelligent quality control is based on a material-machine-information-machine autonomous decision-making interactive system framework, and is used as a rock-roller during earth-rock engineering construction.
  • the system mainly includes: an autonomous decision-making control system for the operating parameters of the roller compactor, multiple unmanned roller compactors 101, a real-time positioning device (ie, RTK-system 103), a wireless communication system 105, and a remote quality monitoring center 104, etc.
  • the unmanned roller compactor 101 can form a cooperative work system, and carry out the unmanned roller compaction construction operation according to the roller compaction operation parameters provided by the independent decision control system of the roller compactor working parameters.
  • the autonomous decision control system for the working parameters of the compactor mainly includes an independent decision model training module for the compactor working parameters and a predictive module for the compactor working parameters.
  • the artificial intelligence algorithm uses real-time compaction detection data to form training data, and then trains the data.
  • a prediction model is generated, and finally the operating parameters of the roller are obtained through the prediction model through the state parameters before rolling.
  • RTK-system 103 including base station, rover 205 and global positioning satellite 106, adopts RTK-GPS (or BDS) system. Based on carrier phase difference technology, it provides cm-level high-precision positioning services and coordinates of the roller operating area for this solution.
  • the construction of the system where the coordinate system of the working area of the compactor is given by the geographical area latitude and longitude boundary point of the working area positioning and measurement tool 107, and then combined with the RTK-GPS (or BDS) system to give the geographical latitude and longitude coordinates of the entire area
  • the precision geographic coordinates are transmitted to the remote quality monitoring center 104 through a wireless communication device.
  • the remote quality monitoring center 104 constructs a 3D operation area range through a server program, and intelligently assigns a straight path for a single operation to each unmanned roller 101.
  • the RTK-system 103 constructs a three-dimensional planar geographic coordinate system 102 of the work area by moving the positioning measurement tool before the roller operation, and then builds the rolling motion track of the unmanned roller 101 in the work area based on this coordinate system.
  • the intelligent path planning algorithm sends control instructions to the roller compactor control device through the data line to achieve unmanned functions, such as automatically controlling the ignition, steering, speed, braking, and U-turn of the roller compactor.
  • the wireless communication system 105 is mainly composed of an on-board communication device, a communication transfer station, and a remote monitoring platform communication system, etc., and provides a real-time high-speed communication service for a single unmanned roller compactor or a plurality of unmanned roller compactor group cooperative rolling operation systems.
  • a real-time communication service for unmanned roller compactors and remote quality monitoring centers in addition, it also provides services for other scenarios that require communication during the rolling construction operation; Transmission of information and instructions, and information communication between the various parts of the on-board automatic control device, multiple unmanned rollers, and other nearby construction equipment.
  • the remote quality monitoring center 104 plans the operation area and the navigation route, receives the operation data and video information of the operation environment of the unmanned roller through the wireless communication system, and sends it to the unmanned roller.
  • Automatic navigation data and instructions man-machine interactive remote operation of unmanned rollers according to receiving emergency processing requests, providing dispatching services for unmanned rollers and cluster cooperative rolling operation systems, and displaying on-site rolling construction operations in real time Store on-site rolling construction data.
  • the unmanned roller compactor 101 has functions such as remote wake-up, hibernation, high-precision RTK-system positioning and navigation, safety obstacle avoidance, independent planning of roller compaction operations, single-machine or multiple-machine cooperative autonomous roller compaction, and automatic detection of construction quality, as shown in Figure 3.
  • the unmanned roller compactor 101 may form multiple working groups for cooperative rolling operations, and is mainly composed of a roller compact body 201, a vibrating roller 202, It consists of intelligent computing device 203, wireless transmission device 204, rover 205, roller control device, on-line compaction quality detection device 210, roller sensing device 211, and anti-collision radar device 212.
  • the intelligent computing device 203 is a high-performance embedded system, which is the most core device in the entire system solution. Its main functions include data analysis of unmanned rollers, unmanned command control, intelligent prediction of the operating parameters of the rollers, and unmanned Operation and maintenance of all schemes of the roller compactor.
  • Intelligent computing device 203 compaction quality online detection device 210, roller control device, anti-collision radar device 212, and roller sensor device 211 are all installed on the roller body 201;
  • the roller control device mainly includes an electric throttle 207, an electric brake device 208, an electric steering wheel 206, and an electric gear control device 209, all of which are located in the cab;
  • the roller sensing device 211 is mounted on the side of the frame of the vibration roller 202.
  • the unmanned roller compactor 101 on the software function module mainly includes a sensing system, a control system, a driving system, a sensing system, a driving system, and a control system to form an on-board automatic control system, and the on-board automatic control device receives remote quality monitoring in real time
  • the automatic navigation instructions and remote driving information sent by the center are compared with the speed, steering angle, and position information detected by the sensor system and RTK-system.
  • the deviation is adjusted by the PID algorithm and transmitted to the electric steering wheel, electric throttle and electric brake system. Move the actuators and control the vibrating roller to complete the desired unmanned rolling operation.
  • the data of the unmanned roller compactor 101 mainly include the position coordinate data of the real-time positioning device, the compaction mass value measured in real time by the compaction quality online detection device 210, the obstacle distance data in the direction of movement of the forward collision avoidance radar device 212, and the crusher.
  • Parameter data such as the vibration frequency amplitude and deflection angle of the rolling wheel generated by the pressing wheel sensing device 211;
  • the unmanned command control will correct the movement status of the roller in real time through the roller control device based on the coordinate data of the roller and the deflection angle data of the roller, and the intelligently planned path data. To ensure that the roller moves straight along the planned single operation.
  • the realization of the intelligent prediction of the working parameters of the roller is achieved through the historical data and the model prediction obtained by remote service training.
  • the training data mainly includes geographic location data (x, y), compaction quality data S n-1 (x, y) of the coordinate distribution of the n-1th compaction, and compaction quality of the coordinate distribution of the nth compaction.
  • S n (x, y) F (S n-1 (x, y), f n-1 (x, y), A n-1 (x, y), V n-1 (x, y))
  • n-1 and n-th rolling effect data is obtained as Sn -1 (x, y) through the online compaction quality detection device and real-time positioning device.
  • S n (x, y) the n-th pass rolling machine operating parameter (f n (x, y) , A n (x, y), V n (x, y)) by rolling the wheel mounted on the sides of the frame
  • This parameter is also the result of the coordinate point adjusted by the control device during the n-th rolling of the coordinate point.
  • each intelligent compaction model is synchronized to the intelligent computing device 203 of the corresponding unmanned roller through the wireless communication system.
  • the intelligent computing device 203 is performing the next position (x i + 1 , y i + 1 )
  • the working parameters of the roller compactor at this point will be adjusted before the n + 1th compaction, that is, the compaction effect index value before the compaction at this point, the specified compaction effect index value, the number of compacted passes, and the geographic coordinates
  • the data were used as model input parameters, and the model was used to predict the vibration frequency, amplitude, and roller speed of the n + 1th roller.
  • the rolled parameter data in this part can be synchronized to the remote quality monitoring center 104 as training data for model update, such as Repeated model iterations can continuously optimize the model, and finally obtain the optimal model data of the earth roller compaction project.
  • model update such as Repeated model iterations can continuously optimize the model, and finally obtain the optimal model data of the earth roller compaction project.
  • the operating area can efficiently achieve the desired compaction effect index value.
  • the remote quality monitoring center 104 will systematically evaluate the overall compaction quality result of the straight road based on the historical data of the straight road, and then give the next schedule of the rolled straight road.
  • the working area flows in sequence until the rolling work is completed in the entire working area.

Abstract

一种土石方智能碾压系统，包括碾压机工作参数自主决策控制系统、无人碾压机及机群协同碾压作业系统、RTK-系统、无线通信系统、远程质量监控中心，基于材料-机器-信息-机器自主决策交互系统框架实现，根据当前压实状态进行下一遍碾压时参数自主决策控制，实现土石方工程精细化填筑施工。

Description

一种土石方智能碾压系统 技术领域
本发明属于土石方工程技术领域，特别涉及一种土石方智能碾压系统，主要用于大坝、公路、铁路、机场、堤坝、桥梁、港口等领域土石方工程填筑碾压质量控制。
背景技术
土石方工程的稳定性由填筑施工质量决定，其具体控制指标有结果控制指标(如压实度K、干密度ρ等)、料源控制指标(如含水量w、级配p等)和施工参数控制指标(如碾压遍数n、行车速度v、振动频率f、行车轨迹等)。土石方压实监控系统是实现对上述三种指标进行全过程监控的有效手段，相关系统已被国内外学者进行了一定的研究并取得了若干成果。其中，碾压机集成压实监控技术(Roller-Integrated Compaction Monitoring，RICM)已经在美国和欧洲的土石方工程中广泛应用。目前国外基于RICM技术研发的压实监控系统主要有美国Caterpillar公司的基于碾压机驱动功率(MDP)的碾压过程监测系统、瑞典Ammann公司的基于土体刚度表征指标K s的ACE(Ammann Compaction Expert)系统、Dynamic与Trimble公司基于加速度表征指标CMV的RICM系统、Rinehart和Mooney基于总谐波失真指标(THD)的RICM系统。连续压实控制系统(Continuous Compaction Control，CCC)是安装在压实装备上能连续采集碾压机操作和性能实时信息的数据采集系统。从某种意义上来说，CCC是RICM的另一种命名。随着技术的发展，BOMAG，AMMANN和Dynapac公司提供了自动反馈控制系统(AFC)，当辊的跳跃模式被确定或达到一个预设的碾压机测量阈值时，系统中的振幅、频率可以自动调节，利用AFC系统对土进行压实可以增加快速压实的机会，改进土特性的均匀性。当RICM系统提 供碾压机频率、振幅、行车速度等的自动反馈控制(AFC)功能时，它往往被称为智能压实系统(Intelligent Compaction，IC)。按照美国联邦高速公路管理局(FHWA)的定义，IC指利用装备有集成测量系统、机载计算机报告系统、基于地图的全球定位系统(GPS)和反馈控制系统的现代振动碾压机对公路材料如土、骨料或沥青路面材料的压实系统。IC技术已经在欧洲和日本实现很多年，该技术在2000年末被引入美国。机器参数的动态控制可以将振动能量应用到欠压区域，防止过度压实和确保土/骨料的均匀压实。尽管IC技术及其系统的应用充满前景，但它们的性能并未得到完全评估。Liu等利用土石坝压实质量实时监控系统，对碾压施工参数(碾压遍数n、压实厚度h、行车轨迹、行车速度v)进行了实时采集，进而分析了含水率w、级配p、压实质量(如压实度K、干密度ρ)和碾压施工参数之间的相关性分析，建立了包含碾压施工参数、含水量、级配、压实质量的多元回归模型，据此模型，结合Kriging空间插值方法，估算了全仓面任意位置处的压实度及全仓面的碾压质量达标率。此外，Liu等还利用填筑碾压质量实时监控技术对混凝土面板堆石坝的压实质量进行了快速质量评估研究，建立了CV值和含水量、级配为变量的土石坝压实质量评估模型，结合Kriging空间插值方法，提出了基于CV值的土石坝填筑材料压实质量全工作面的快速评估方法，实现了全工作面压实质量的快速评估。
在国内，关于土石方压实监控系统方面的研究也取得了一定成果。经过对加速度传感器采集到的数据与土的压实度之间关系的研究，张润利等研制了加速度计。邓学欣等提出通过检测振动轮振动加速度间接反映土壤压实状况的压实度自动检测原理，并开发了响应检测系统。范云等开展了填土压实质量检测及机载压实集成系统应用研究。对于工程量较大、工序复杂的面板堆石坝，基于GPS技术，黄声享等研制了可实时监控碾压遍数、压实厚度和行车速度的监控系统。为了解决粗粒土路基压实质量较难控制的问题，基于连续检测路基结构抗力变化的方法，徐光辉等提出了路基压实质量连续动态监控技术。针对南水北调中线工程干渠渠道大部分为高填方工程，土石方工程量巨大，工期紧等 问题，李斌等研制开发了南水北调中线一期工程高填方段碾压施工质量实时监控系统，并在工程实践中得到应用。于子忠和黄增刚开展了智能压实过程控制系统在水利水电工程中的试验性应用研究。针对南水北调工程中高填方段的施工特点，余洋等开发了相应的填筑施工质量监控系统，并在该工程中进行了应用。曾怀恩和刘金平对土石坝碾压实时监控系统进行了模拟试验研究。刘东海等以反映碾压机做功的坝料单位体积压实功(E)和振动轮的加速度谐波失真量(THD 0)作为堆石坝压实质量的实时监测指标，研制开发了堆石坝压实质量实时监测系统，对于堆石坝的压实质量进行了实时监测与快速评估。针对传统的以随机取样的干密度实验检测坝体填筑质量的方法不能全面反映坝体实际碾压质量，影响坝体质量评价可靠性的问题，王晓玲等利用压实质量实时监控系统对于堆石坝的碾压质量进行了二元耦合评价研究。基于实时采集到的碾压施工参数数据，刘东海和王东烽进行了不同碾压施工参数(如行车速度v、压实厚度h)与压实质量评估指标(如压实度K、干密度ρ)的相关性分析，针对填筑材料的压实质量模型得以建立，进而结合地质统计方法提出一种可评估仓面上任意位置的压实质量及全工作面的压实质量达标率的快速评估方法。
除了RICM/CCC/IC系统之外，基于GPS技术和自动控制技术的自动导航控制系统目前已取得一定研究成果。在国外，相关研究主要集中在道路施工方面。在国内，长沙矿山研究院于1995年研制了具有遥控功能的无人碾压振动压路机，并应用于矿山道路建设。2000年，湘潭江麓机械公司与国科大联合研制了W1102DZ型高性能无人碾压振动压路机，在现场测试时能够自主完成点火、起步、变速、转向、倒车、停车等基本操作。针对粒径分布范围较广的土石料，刘天云等研制了一种水利施工振动自动驾驶系统，该系统可有效解决相邻作业面间重复、漏碾、交叉碾压等问题。此外，在农业领域，国内也取得了一定研究成果。华南农业大学与雷沃重工研制了基于RTK-GPS技术的自动导航控制系统，解决了农田播种“播行不直、接行不准”的瓶颈问题。基于XDNZ630型睡到插秧机、RTK-GPS技术和自动控制技术，伟利国等研制了插 秧机自动导航转向系统，该系统具有自动对行导航和地头转向的功能。
以上国内外最新的研究成果表明，国外的土石方压实监控技术主要应用于公路建造中，在其他领域缺乏应用研究，且技术水平仍停留于自动反馈控制阶段，土石方工程碾压质量的控制不仅对操作人员要求较高，而且仍由人来完成决策实施，而国内相应的技术仅仅发展到对土石方工程填筑质量进行实时碾压施工参数监控，这种监控仅仅是针对碾压遍数、行车速度、压实厚度、碾压机振动状态等的监控且偏重于对压实质量进行监测，而疏于对压实质量的控制，对土石方工程填筑碾压质量的监控基本上还是通过碾压过程中对碾压施工参数进行实时监控、碾压作业完成后建立压实质量评估模型预测整个施工区域碾压层的压实质量，这样很容易在材料碾压过程中产生过压或欠压的情况，使得压实均匀性得不到保证。除此之外，无人驾驶碾压技术等也是处于自动控制阶段并以人来决策评估填筑材料的压实质量。总的来说，目前的研究仍基于材料-机器-信息-人决策交互系统框架，仍基于人工决策对填筑料碾压质量进行控制，对土石方工程填筑碾压质量的控制并非实现智能控制，压实质量得不到有效保证，碾压作业精确性、填筑料的压实均匀性、现场施工作业效率得不到保障。
发明内容
为了克服上述现有技术的缺点，本发明的目的在于提供一种土石方智能碾压系统，该系统基于材料-机器-信息-机器自主决策交互系统框架，将智能决策控制方法应用于压实质量监控系统中，实现真正的智能压实，该系统可用于土石方工程填筑施工质量控制，作为土石方工程施工过程中碾压质量智能化控制的一种有效解决方案，可以对碾压质量实现具有机器自主决策功能的智能压实控制，以填筑材料压实质量为控制目标，以最高效的控制手段使得填筑料达到压实质量(如压实度K)的目标期望值，而不会出现过压、欠压或压实均匀性差的情况。该系统主要包括无人驾驶碾压系统和碾压机工作参数自主决策控制系统，实现压实参数机器自主决策控制，以实现智能高效的碾压。
为了实现上述目的，本发明采用的技术方案是：
一种土石方智能碾压系统，基于材料-机器-信息-机器自主决策交互系统框架实现，包括：
碾压机工作参数自主决策控制系统，通过人工智能算法，利用实时压实情况检测数据组成训练数据，然后通过数据训练产生预测模型，最后通过碾压前的状态参数经过预测模型得到碾压机作业参数，所述实时压实情况检测数据包括该位置的压实效果指标值S(x,y)和碾压机工作参数，所述碾压机工作参数包括振动频率f(x,y)，振幅A(x,y)和行车速度V(x,y)，还可包括激振力等；
无人碾压机及其机群协同碾压作业系统，其中无人碾压机根据碾压作业参数进行无人碾压施工作业；
RTK-系统，为无人碾压机提供高精度的定位导航服务；
无线通信系统，为单台无人碾压机或多台无人碾压机机群协同碾压作业系统提供实时高速通信服务，为无人碾压机与远程质量监控中心提供实时通信服务；此外，也为碾压施工作业过程中其他需要通信的场景提供服务；
远程质量监控中心，依据建设土石方工程的三维数字模型人机交互规划作业区域和导航路线，通过无线通信系统接收无人碾压机的作业数据及作业环境视频信息，向无人碾压机发送自动导航数据与指令，根据接收紧急处理请求情况人机交互远程操纵无人碾压机，为无人碾压机及机群协同碾压作业系统提供调度服务，并显示现场碾压施工作业情况，实时存储现场碾压施工作业数据。
所述碾压机工作参数自主决策控制系统主要包含碾压机工作参数自主决策模型训练模块和碾压机工作参数预测模块，其中，碾压机工作参数自主决策模型训练模块通过人工智能算法(如机器学习算法)模型进行数据训练产生预测模型，碾压机工作参数预测模块利用输入参数和周期训练的模型运算，得到下一个局部区域将需要的最大效率的碾压机工作参数，制定压实策略，实现压实过程智能化控制。
碾压机工作参数自主决策模型训练模块和碾压机工作参数预测模块又可称为碾压机智能压实模型模块：智能压实模型主要建立压实效果与碾压参数的关 系，并根据当前压实状况，进行碾压参数优化，制定压实策略，实现压实过程智能化控制。
所述碾压机工作参数自主决策模型训练模块利用实时定位装置、压实质量在线检测装置、传感器装置得到的数据，包括该位置坐标(x,y)和实时压实情况检测数据，组成训练数据，通过坐标数据(x,y)，当前已经碾压遍数(n-1)(x,y)，碾压前压实效果指标值S n-1(x,y)、第n-1遍碾压机工作参数、振动频率f n-1(x,y)、振幅A n-1(x,y)以及速度V n-1(x,y)等数据组成训练模型的输入数据，碾压后压实效果指标值S n(x,y)作为训练模型的输出数据，通过机器学习算法模型进行数据训练，得到压实效果与碾压工作参数的关系，即：
S n(x,y)＝F(S n-1(x,y),f n-1(x,y),A n-1(x,y),V n-1(x,y))
所述碾压机工作参数预测模块在无人碾压机实时作业时，将实时的从实时定位装置、压实质量在线检测装置以及传感器装置等获得当前压实状态S n-1(x,y)，以达到规定压实效果指标S E(x,y)的效率最高为目标进行碾压参数优化，得到智能压实模型的输出：
(f n(x,y),A n(x,y),V n(x,y))＝F(S n-1(x,y),S E(x,y))
f n(x,y),A n(x,y),V n(x,y)即下一遍碾压作业的碾压机振动频率、碾压机振幅和碾压机车速，实现碾压作业参数自主决策控制。
所述碾压机工作参数自主决策控制系统还用于土石方工程填筑碾压质量评估、质量反馈控制(QC)和质量验收(QA)，即碾压机工作参数自主决策控制系统依照设计填筑压实质量的控制标准，将获取的压实效果指标值通过压实质量评估模型计算出相应的设计填筑压实质量的控制指标值，进而计算该值是否符合设计填筑压实质量的控制标准，因而可实现土石方工程填筑碾压质量评估，通过评估结果可利用碾压机工作参数自主决策控制系统对压实质量作反馈控制和质量验收，具有机器自主决策进行质量评估、质量反馈控制(QC)和质量验收(QA)的智能化特点。
所述无人碾压机具备远程唤醒、休眠、高精度RTK-系统定位导航、安全 避障、自主规划碾压作业、单机或多机协作自主碾压与自动检测施工质量等功能，主要包含传感系统、控制系统、驱动系统和碾压机本体，传感系统、驱动系统及控制系统组成机载自动控制系统，机载自动控制装置实时接收远程质量监控中心发来的自动导航指令、遥控驾驶信息，并与传感系统、RTK-系统检测的速度、转向角度、位置等信息比较，偏差经PID算法调整后传送至电动方向盘、电动油门以及电动刹车制动等执行元件，控制振动碾压机完成期望的无人驾驶碾压作业。
无人碾压机具体可由无人碾压机控制模块实现控制：
该模块主要基于实时定位装置、无线通信系统、智能运算装置、碾压机操控装置来共同协作实现，首先实时定位装置在碾压机作业前通过移动定位测量工具构建作业区域的三维平面地理坐标体系，然后智能运算装置将基于此三维平面地理坐标体系构建出无人碾压机在作业区域的碾压运动轨迹。在碾压机作业时，将通过智能运算设备智能路径规划算法通过数据线向碾压机操控装置发送控制指令来实现无人驾驶的功能，如自动控制碾压的点火、转向、速度、制动、掉头等。
所述RTK(Real-time kinematic，实时动态)-系统为RTK-GPS系统或RTK-BDS系统，基于载波相位差分技术和GPS(或BDS)经纬度数据实现cm级的高精度定位，RTK-系统获取高精度的全球经纬度数据，构建三维平面地理坐标体系，然后通过映射算法来转换成高精度的平面地理坐标体系。其主要由GPS或BDS接收机、RTK-GPS或BDS基准站、现场流动站、卫星等共同组成，实时提供精确的无人碾压机的空间位置信息、行车速度信息和其他运行状态信息，实时计算车体与最优路径之间的距离偏差和角度偏差，从而计算出调整转向角，为无人碾压机提供高精度定位导航服务。
所述传感系统包括防撞雷达装置和碾压轮传感装置，所述控制系统包括智能运算装置、压实质量在线检测装置和碾压机操控装置，防撞雷达装置安装在碾压机的最前方，实时监测碾压机行驶方向的障碍物的空间距离，当距离运动 方向的障碍距离达到预警阈值时候，无人碾压机将采取智能避让策略。所述压实质量在线检测装置由传感器、传感器安装装置、数据采集卡、数据处理与分析装置构成，属于填筑施工碾压质量机载测量技术，通过测量碾压过程中与填筑料压实质量相关的信号，实时在线获取填筑材料的压实效果指标即RICMMV、ICMV或SCV，并通过压实效果指标来表征填筑料的压实质量，不同的压实效果指标均可用于表征填筑料压实质量(如刚度、强度、变形、沉降、温度、密度或压实度)；所述压实质量在线检测装置周期性地检测当前位置当前时刻的压实效果指标值，当前的位置信息通过碾压机上的定位装置获得，压实效果指标值通过无线装置传输到远程质量监控中心，压实效果指标值用可视化效果展示。
所述无线通信系统主要由机载通信装置、通信中转站、远程监控平台通信系统组成，实现无人碾压机与远程质量监控中心之间的信息和指令的传输，并实现机载自动控制装置各部分之间、多台无人碾压机之间，以及与附近其他施工设备间的信息通信。
所述无人碾压机的机群协同碾压作业系统基于无人碾压机、远程质量监控中心多机调度系统、工程场地路径规划算法和RTK-GPS或RTK-BDS导航技术构建，远程质量监控中心制定导航线路并通过无线通信系统向无人碾压机自动控制装置提供，根据碾压机工作参数自主决策控制系统得出的碾压机实时工作参数控制无人碾压机的动作执行，进行多机协作智能碾压作业以及自动碾压，所述远程质量监控中心多机调度系统根据现场实际情况给出无人碾压机调度实施方案，实现多台无人碾压机的有效调度，为无人碾压机机群协同碾压作业提供基础服务；所述工程场地路径规划算法为无人碾压机提供最优的碾压作业和避障导航路径。
所述远程质量监控中心为作业区域的大数据数字化可视化平台和与无人碾压机交互平台，实时显示碾压施工作业现场情况，在紧急或特殊情况下进行人机交互远程操控无人碾压机作业，其中碾压施工作业现场情况包含现场填筑料 碾压质量信息(如压实质量连续表征指标信息)、碾压施工参数信息(如行车速度、碾压遍数、压实厚度、振动频率)和现场环境视频信息。其可实现对作业区域的碾压质量评估结果3D化的展示，同时将作业区域的所有运行数据进行智能分析、建模和存储到数据库中，然后周期的更新预测模型和通过无线传输到无人碾压机的智能设备装置中，以及对无人碾压机的紧急制动、碾压机召回到基地区域等控制交互。
与现有技术相比，本发明具有智能、高效、实时、控制精度高等特点，可根据当前压实状态进行下一遍碾压作业时碾压参数实时自主决策控制，实现碾压过程机器自主决策控制，适用于土石方工程填筑施工质量控制，可通过对表征填筑料压实质量的连续压实指标值进行快速检测，确定填筑料压实质量并计算压实模型参数，利用压实模型进行碾压机工作参数优化，提出压实决策最优方案，以控制碾压机工作参数。本发明可以应用于水利、公路、铁路、机场、港口等土石方工程填筑施工质量控制，不仅可有效控制碾压施工质量，也可有效提高碾压施工效率，实现碾压过程的智能压实控制，实现土石方工程填筑施工精细化管理
附图说明
图1为本发明提供的智能压实决策系统的框架图。
图2为本发明提供的智能压实决策系统的原理示意图。
图3为本发明提供的智能压实决策系统的软件系统流程图。
图4为本发明提供的无人碾压机的一个实施例的结构图。
具体实施方式
下面结合具体实施例和附图对本发明进行详细说明。以下实施例将有助于本领域的技术人员进一步理解本发明，但不以任何形式限制本发明。应当指出的是，对本领域的普通技术人员来说，在不脱离本发明构思的前提下，还可以做出若干变化和改进。这些都属于本发明的保护范围。
本发明提供一种土石方智能碾压系统，可用于土石方工程填筑施工质量控 制，如图1所示，该系统基于材料-机器-信息-机器自主决策交互系统框架，作为土石方工程施工过程中碾压质量智能化控制的一种有效解决方案。
如图2所示，该系统主要包括：碾压机工作参数自主决策控制系统、多台无人碾压机101、实时定位装置(即RTK-系统103)、无线通信系统105、远程质量监控中心104等，无人碾压机101可组成协同工作系统，根据碾压机工作参数自主决策控制系统提供的碾压作业参数进行无人碾压施工作业。
碾压机工作参数自主决策控制系统主要包含碾压机工作参数自主决策模型训练模块和碾压机工作参数预测模块，通过人工智能算法，利用实时压实情况检测数据组成训练数据，然后通过数据训练产生预测模型，最后通过碾压前的状态参数经过预测模型得到碾压机作业参数。
RTK-系统103，包括基站、流动站205以及全球定位卫星106，采取RTK-GPS(或BDS)系统，基于载波相位差分技术为该方案提供cm级高精度的定位服务和碾压机作业区域坐标系统的构建，其中碾压机作业区域坐标系统是由作业区域定位测量工具107给出区域的地理经纬度边界点，然后结合RTK-GPS(或BDS)系统给出整个区域的地理经纬度坐标，该高精度地理坐标通过无线通信装置传输到远程质量监控中心104，远程质量监控中心104通过服务器程序构建出3D作业区域范围，并给每台无人碾压机101智能分配单次作业的直道路径。
RTK-系统103在碾压机作业前通过移动定位测量工具构建作业区域的三维平面地理坐标体系102，然后基于此坐标体系构建出无人碾压机101在作业区域的碾压运动轨迹。在碾压机作业时，智能路径规划算法通过数据线向碾压机操控装置发送控制指令来实现无人驾驶的功能，如自动控制碾压的点火、转向、速度、制动、掉头等。
无线通信系统105主要由机载通信装置、通信中转站、远程监控平台通信系统等组成，为单台无人碾压机或多台无人碾压机机群协同碾压作业系统提供实时高速通信服务，为无人碾压机与远程质量监控中心提供实时通信服务；此 外，也为碾压施工作业过程中其他需要通信的场景提供服务；可实现无人碾压机与远程质量监控中心之间的信息和指令的传输，并实现机载自动控制装置各部分之间、多台无人碾压机之间，以及与附近其他施工设备间的信息通信。
远程质量监控中心104，依据建设土石方工程的三维数字模型人机交互规划作业区域和导航路线，通过无线通信系统接收无人碾压机的作业数据及作业环境视频信息，向无人碾压机发送自动导航数据与指令，根据接收紧急处理请求情况人机交互远程操纵无人碾压机，为无人碾压机及机群协同碾压作业系统提供调度服务，并显示现场碾压施工作业情况，实时存储现场碾压施工作业数据。
无人碾压机101具备远程唤醒、休眠、高精度RTK-系统定位导航、安全避障、自主规划碾压作业、单机或多机协作自主碾压与自动检测施工质量等功能，图3所示为本发明提供的无人碾压机101的一个实施例的结构图，无人碾压机101可构成多台作业群协同碾压作业，主要由碾压机机体201、振动碾压轮202、智能运算装置203、无线传输装置204、流动站205、碾压机操控装置、压实质量在线检测装置210、碾压轮传感装置211、防撞雷达装置212等组成。
智能运算装置203为高性能的嵌入式系统，是整个系统方案中最核心的装置，主要功能有无人碾压机的数据分析、无人驾驶指令控制、碾压机工作参数智能预测、无人碾压机所有方案附加装置的运行维护等。
智能运算装置203、压实质量在线检测装置210、碾压机操控装置、防撞雷达装置212、碾压轮传感装置211均安装在碾压机机体201上；
碾压机操控装置主要包括电动油门207、电动制动装置208、电动方向盘206以及电动档位控制装置209，均设置在驾驶室内；
碾压轮传感装置211装在振动碾压轮202的框架侧面。
无人碾压机101在软件功能模块上，主要包含传感系统、控制系统、驱动系统，传感系统、驱动系统及控制系统组成机载自动控制系统，机载自动控制装置实时接收远程质量监控中心发来的自动导航指令、遥控驾驶信息，并与传 感系统、RTK-系统检测的速度、转向角度、位置等信息比较，偏差经PID算法调整后传送至电动方向盘、电动油门以及电动刹车制动等执行元件，控制振动碾压机完成期望的无人驾驶碾压作业。
无人碾压机101的数据主要有实时定位装置的位置坐标数据、压实质量在线检测装置210所实时测得压实质量数值、前方防撞雷达装置212的运动方向的障碍物距离数据、碾压轮传感装置211产生的碾压轮振动频率幅度和偏转角等参数数据；
无人驾驶指令控制将根据获得到实时碾压机的坐标数据和碾压轮的偏转角数据，再结合智能规划的路径数据通过碾压机操控装置实时进行对碾压轮的运动状态进行修正，以保证碾压机沿着规划的单次作业直道运动。
碾压机工作参数智能预测的实现是通过历史数据由远程服务训练得到模型预测来实现。训练数据主要包括地理位置数据(x,y)、第n-1遍碾压的坐标分布的压实质量数据S n-1(x,y)，第n遍碾压的坐标分布的压实质量数据S n(x,y)、实时的碾压机工作参数(振动频率f n(x,y)、振幅A n(x,y)、行车速度V n(x,y)、当前的碾压次数n(x,y)等，这些组合数据通过无线传输装置实时传输到远程质量监控中心进行数据分析、可视化展现、周期性的预测模型更新，得到新压实预测模型：
S n(x,y)＝F(S n-1(x,y),f n-1(x,y),A n-1(x,y),V n-1(x,y))
然后以达到规定压实效果指标S E(x,y)的效率最高为目标进行碾压参数优化，得到智能压实模型的输出：
(f n(x,y),A n(x,y),V n(x,y))＝F(S n-1(x,y),S E(x,y))
该智能压实自主决策系统实现流程如图4所示，其中第n-1和n遍的碾压效果数据通过压实质量在线检测装置和实时定位装置获得为S n-1(x,y)和S n(x,y)，第n遍碾压机工作参数(f n(x,y),A n(x,y),V n(x,y))通过安装在碾压轮侧面框架的碾压轮传感装置获得，此参数也是该坐标点在第n遍碾压时候通过操控装置调节的结果。
每次智能压实模型的输出结果通过无线通信系统同步到相应的无人碾压机的智能运算装置203中，智能运算装置203在进行下一位置点(x i+1,y i+1)第n+1遍碾压前将对该点的碾压机工作参数调整，即通过把该点碾压前的压实效果指标值、规定压实效果指标值、已碾压遍数、地理坐标数据作为模型输入参数，利用模型预测得第n+1遍碾压机的振动频率、振幅、碾压轮速度。当碾压过该点(x i+1,y i+1)第n+1遍后，此部分的碾压后的参数数据又可以作为训练数据同步到远程质量监控中心104进行模型更新，如次反复进行模型迭代可以不断的优化模型、最后得到此土方碾压工程的最优模型数据，通过此模型可以实现作业区域高效的达到期望压实效果指标值。当碾压机行驶到图2直道的一端，远程质量监控中心104将通过该直道的历史数据进行系统评估出该直道的整体压实质量结果，然后给出下一次的碾压直道的调度。作业区依次流程直至整个作业区域完成碾压工程。

Claims (10)

  1. 一种土石方智能碾压系统，其特征在于，基于材料-机器-信息-机器自主决策交互系统框架实现，包括：
    碾压机工作参数自主决策控制系统，通过人工智能算法，利用实时压实情况检测数据组成训练数据，然后通过数据训练产生预测模型，最后通过碾压前的状态参数经过预测模型得到碾压机作业参数，所述实时压实情况检测数据包括该位置的压实效果指标值S(x,y)和碾压机工作参数，所述碾压机工作参数包括振动频率f(x,y)，振幅A(x,y)和行车速度V(x,y)；
    无人碾压机及其机群协同碾压作业系统，其中无人碾压机根据碾压作业参数进行无人碾压施工作业；
    RTK-系统，为无人碾压机提供定位导航服务；
    无线通信系统，为单台无人碾压机或多台无人碾压机机群协同碾压作业系统提供实时通信服务，为无人碾压机与远程质量监控中心提供实时通信服务；
    远程质量监控中心，依据建设土石方工程的三维数字模型人机交互规划作业区域和导航路线，通过无线通信系统接收无人碾压机的作业数据及作业环境视频信息，向无人碾压机发送自动导航数据与指令，根据接收紧急处理请求情况人机交互远程操纵无人碾压机，为无人碾压机及机群协同碾压作业系统提供调度服务，并显示现场碾压施工作业情况，实时存储现场碾压施工作业数据。
  2. 根据权利要求1所述土石方智能碾压系统，其特征在于，所述碾压机工作参数自主决策控制系统主要包含碾压机工作参数自主决策模型训练模块和碾压机工作参数预测模块，其中，碾压机工作参数自主决策模型训练模块通过人工智能算法模型进行数据训练产生预测模型，碾压机工作参数预测模块利用输入参数和周期训练的模型运算，得到下一个局部区域将需要的最大效率的碾压机工作参数，制定压实策略，实现压实过程智能化控制。
  3. 根据权利要求2所述土石方智能碾压系统，其特征在于，所述碾压机 工作参数自主决策模型训练模块利用实时定位装置、压实质量在线检测装置、传感器装置得到的数据，包括该位置坐标(x,y)和实时压实情况检测数据，组成训练数据，通过坐标数据(x,y)，当前已经碾压遍数(n-1)(x,y)，碾压前压实效果指标值S N-1(x,y)、第n-1遍碾压机工作参数、振动频率f n-1(x,y)、振幅A n-1(x,y)以及速度V n-1(x,y)组成训练模型的输入数据，碾压后压实效果指标值S n(x,y)作为训练模型的输出数据，通过机器学习算法模型进行数据训练，得到压实效果与碾压工作参数的关系，即：
    S n(x,y)＝F(S n-1(x,y),f n-1(x,y),A n-1(x,y),V n-1(x,y))
    所述碾压机工作参数预测模块在无人碾压机实时作业时，将实时的从实时定位装置、压实质量在线检测装置以及传感器装置获得当前压实状态S n-1(x,y)，以达到规定压实效果指标S E(x,y)的效率最高为目标进行碾压参数优化，得到智能压实模型的输出：
    (f n(x,y),A n(x,y),V n(x,y))＝F(S n-1(x,y),S E(x,y))
    f n(x,y),A n(x,y),V n(x,y)即下一遍碾压作业的碾压机振动频率、碾压机振幅和碾压机车速，实现碾压作业参数自主决策控制。
  4. 根据权利要求2所述土石方智能碾压系统，其特征在于，所述碾压机工作参数自主决策控制系统还用于土石方工程填筑碾压质量评估、质量反馈控制和质量验收，即碾压机工作参数自主决策控制系统依照设计填筑压实质量的控制标准，将获取的压实效果指标值通过压实质量评估模型计算出相应的设计填筑压实质量的控制指标值，进而计算该值是否符合设计填筑压实质量的控制标准，因而可实现土石方工程填筑碾压质量评估，通过评估结果可利用碾压机工作参数自主决策控制系统对压实质量作反馈控制和质量验收。
  5. 根据权利要求1所述土石方智能碾压系统，其特征在于，所述无人碾压机具备远程唤醒、休眠、高精度RTK-系统定位导航、安全避障、自主规划碾压作业、单机或多机协作自主碾压与自动检测施工质量功能，主要包含传感系统、控制系统、驱动系统和碾压机本体，传感系统、驱动系统及控制系统组 成机载自动控制系统，机载自动控制装置实时接收远程质量监控中心发来的自动导航指令、遥控驾驶信息，并与传感系统、RTK-系统检测的速度、转向角度、位置信息比较，偏差经PID算法调整后传送至执行元件，控制振动碾压机完成期望的无人驾驶碾压作业。
  6. 根据权利要求5所述土石方智能碾压系统，其特征在于，所述RTK-系统为RTK-GPS系统或RTK-BDS系统，主要由GPS或BDS接收机、RTK-GPS或BDS基准站、现场流动站、卫星共同组成，实时提供精确的无人碾压机的空间位置信息、行车速度信息和其他运行状态信息，实时计算车体与最优路径之间的距离偏差和角度偏差，从而计算出调整转向角，为无人碾压机提供高精度定位导航服务。
  7. 根据权利要求5所述土石方智能碾压系统，其特征在于，所述传感系统包括防撞雷达装置和碾压轮传感装置，所述控制系统包括智能运算装置、压实质量在线检测装置和碾压机操控装置，所述压实质量在线检测装置由传感器、传感器安装装置、数据采集卡、数据处理与分析装置构成，通过测量碾压过程中与填筑料压实质量相关的信号，实时在线获取填筑材料的压实效果指标即RICMMV、ICMV或SCV，并通过压实效果指标来表征填筑料的压实质量，不同的压实效果指标均可用于表征填筑料压实质量；所述压实质量在线检测装置(210)周期性地检测当前位置当前时刻的压实效果指标值，当前的位置信息通过碾压机上的定位装置获得，压实效果指标值通过无线装置传输到远程质量监控中心，压实效果指标值用可视化效果展示。
  8. 根据权利要求1所述土石方智能碾压系统，其特征在于，所述无线通信系统主要由机载通信装置、通信中转站、远程监控平台通信系统组成，实现无人碾压机与远程质量监控中心之间的信息和指令的传输，并实现机载自动控制装置各部分之间、多台无人碾压机之间，以及与附近其他施工设备间的信息通信。
  9. 根据权利要求1所述土石方智能碾压系统，其特征在于，所述无人碾 压机的机群协同碾压作业系统基于无人碾压机、远程质量监控中心多机调度系统、工程场地路径规划算法和RTK-GPS或RTK-BDS导航技术构建，远程质量监控中心制定导航线路并通过无线通信系统向无人碾压机自动控制装置提供，根据碾压机工作参数自主决策控制系统得出的碾压机实时工作参数控制无人碾压机的动作执行，进行多机协作智能碾压作业以及自动碾压，所述远程质量监控中心多机调度系统根据现场实际情况给出无人碾压机调度实施方案，实现多台无人碾压机的有效调度，为无人碾压机机群协同碾压作业提供基础服务；所述工程场地路径规划算法为无人碾压机提供最优的碾压作业和避障导航路径。
  10. 根据权利要求9所述土石方智能碾压系统，其特征在于，所述远程质量监控中心实时显示碾压施工作业现场情况，在紧急或特殊情况下进行人机交互远程操控无人碾压机作业，其中碾压施工作业现场情况包含现场填筑料碾压质量信息、碾压施工参数信息和现场环境视频信息。
PCT/CN2018/094460 2018-07-03 2018-07-04 一种土石方智能碾压系统 WO2020006698A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201810720255.3 2018-07-03
CN201810720255.3A CN108646755A (zh) 2018-07-03 2018-07-03 一种土石方智能碾压系统

Publications (1)

Publication Number Publication Date
WO2020006698A1 true WO2020006698A1 (zh) 2020-01-09

Family

ID=63750378

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2018/094460 WO2020006698A1 (zh) 2018-07-03 2018-07-04 一种土石方智能碾压系统

Country Status (2)

Country Link
CN (1) CN108646755A (zh)
WO (1) WO2020006698A1 (zh)

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112051817A (zh) * 2020-09-04 2020-12-08 天津大学 一种土石坝智能摊铺监控系统及其监控方法
CN112391908A (zh) * 2020-12-04 2021-02-23 天津大学 一种无人驾驶碾压机控制系统
CN113607272A (zh) * 2021-07-30 2021-11-05 清华大学 一种碾压机工作状态的监控方法及系统
CN113671948A (zh) * 2021-07-27 2021-11-19 北京科技大学 一种土石方工程无人碾压机机群协同智能作业控制方法
CN115564396A (zh) * 2022-10-25 2023-01-03 中国铁道科学研究院集团有限公司铁道建筑研究所 一种融合5g通讯技术的路基智能作业系统
EP4332302A1 (de) * 2022-08-29 2024-03-06 MOBA Mobile Automation AG Verdichter

Families Citing this family (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109468918B (zh) * 2018-10-31 2021-06-04 北京龙马智行科技有限公司 一种路基路面智能压实决策系统
CN109542105A (zh) * 2018-12-28 2019-03-29 北京龙马智行科技有限公司 一种路基路面智能无人碾压系统
CN109881566A (zh) * 2019-03-19 2019-06-14 中国铁道科学研究院集团有限公司铁道建筑研究所 振动压路机智能调频调幅碾压方法
CN110331639B (zh) * 2019-07-08 2021-07-30 长安大学 一种可自主作业智能压路机系统
CN110766226A (zh) * 2019-10-24 2020-02-07 桂林航天工业学院 一种基于智能建模和智能算法的平地机远程作业调度法
CN111445170A (zh) * 2020-04-30 2020-07-24 天津大学 一种无人碾压机群自主作业系统及方法
CN111459172B (zh) * 2020-05-20 2023-10-31 中国北方车辆研究所 围界安防无人巡逻车自主导航系统
CN112434912A (zh) * 2020-11-02 2021-03-02 中建三局第二建设工程有限责任公司 一种基于机械自动化的沙漠地带场平固沙方法
CN113158558B (zh) * 2021-04-02 2023-02-24 哈尔滨理工大学 一种高速铁路路基连续压实分析方法、装置及分析仪
CN113512924B (zh) * 2021-08-16 2022-04-01 河北工业大学 一种外挂式压路机无人辅助驾驶系统
CN114197275B (zh) * 2022-01-21 2023-04-25 山推工程机械股份有限公司 一种压路机的控制系统及方法
CN115491952A (zh) * 2022-09-30 2022-12-20 湖南交通国际经济工程合作有限公司 一种公路路基路面智慧压实监控系统
CN116383697B (zh) * 2023-03-29 2024-02-06 四川省交通建设集团有限责任公司 一种基于双门控循环网络的沥青路面压实程度预测方法
CN116485063B (zh) * 2023-06-20 2023-08-18 成都工业职业技术学院 一种基于大数据的无人驾驶碾压机群控制方法及装置
CN117744858A (zh) * 2023-12-04 2024-03-22 南京交科数智科技发展有限公司 一种基于大数据分析的路面压实度实时预测系统及方法

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2000008535A1 (en) * 1998-08-06 2000-02-17 Caterpillar Inc. Method and apparatus for establishing a perimeter defining an area to be traversed by a mobile machine
CN102289716A (zh) * 2011-06-18 2011-12-21 合肥工业大学 智能压路机最佳工作参数的神经网络建模方法
CN105002810A (zh) * 2015-06-01 2015-10-28 清华大学 一种智能碾压机器人
CN105137997A (zh) * 2015-09-22 2015-12-09 清华大学 水利施工振动碾压机自动驾驶系统与方法
CN107761701A (zh) * 2017-10-24 2018-03-06 上海交通大学 用于土石方碾压的无人驾驶智能振动碾压机及系统

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
GB9504345D0 (en) * 1995-03-03 1995-04-19 Compaction Tech Soil Ltd Method and apparatus for monitoring soil compaction
CN202298427U (zh) * 2011-09-19 2012-07-04 刘振燕 振动碾压机智能轨迹控制系统
CN104423333A (zh) * 2013-08-23 2015-03-18 西安发威电子科技有限公司 一种压路机施工监控系统
CN205353747U (zh) * 2016-01-18 2016-06-29 湖南致同工程科技有限公司 用于多台压路机协同作业的系统

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2000008535A1 (en) * 1998-08-06 2000-02-17 Caterpillar Inc. Method and apparatus for establishing a perimeter defining an area to be traversed by a mobile machine
CN102289716A (zh) * 2011-06-18 2011-12-21 合肥工业大学 智能压路机最佳工作参数的神经网络建模方法
CN105002810A (zh) * 2015-06-01 2015-10-28 清华大学 一种智能碾压机器人
CN105137997A (zh) * 2015-09-22 2015-12-09 清华大学 水利施工振动碾压机自动驾驶系统与方法
CN107761701A (zh) * 2017-10-24 2018-03-06 上海交通大学 用于土石方碾压的无人驾驶智能振动碾压机及系统

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112051817A (zh) * 2020-09-04 2020-12-08 天津大学 一种土石坝智能摊铺监控系统及其监控方法
CN112051817B (zh) * 2020-09-04 2024-04-09 天津大学 一种土石坝智能摊铺监控系统及其监控方法
CN112391908A (zh) * 2020-12-04 2021-02-23 天津大学 一种无人驾驶碾压机控制系统
CN113671948A (zh) * 2021-07-27 2021-11-19 北京科技大学 一种土石方工程无人碾压机机群协同智能作业控制方法
CN113671948B (zh) * 2021-07-27 2023-08-22 北京科技大学 一种土石方工程无人碾压机机群协同智能作业控制方法
CN113607272A (zh) * 2021-07-30 2021-11-05 清华大学 一种碾压机工作状态的监控方法及系统
EP4332302A1 (de) * 2022-08-29 2024-03-06 MOBA Mobile Automation AG Verdichter
CN115564396A (zh) * 2022-10-25 2023-01-03 中国铁道科学研究院集团有限公司铁道建筑研究所 一种融合5g通讯技术的路基智能作业系统

Also Published As

Publication number Publication date
CN108646755A (zh) 2018-10-12

Similar Documents

Publication Publication Date Title
WO2020006698A1 (zh) 一种土石方智能碾压系统
CN109356003B (zh) 一种路基路面智能碾压系统
Zhang et al. Unmanned rolling compaction system for rockfill materials
CN109468918B (zh) 一种路基路面智能压实决策系统
CN105002810B (zh) 一种智能碾压机器人
Xu et al. Adaptive quality control and acceptance of pavement material density for intelligent road construction
CN109542105A (zh) 一种路基路面智能无人碾压系统
Zhang et al. Intelligent rolling compaction system for earth-rock dams
US10990245B2 (en) Mobile process management tool for paving operations
CN109630154A (zh) 一种用于隧道掘进的掘进机器人及远程移动终端指挥系统
Zhong et al. Technology and application of real-time compaction quality monitoring for earth-rockfill dam construction in deep narrow valley
CN108254068B (zh) 一种填筑工程压实质量的振动能量在线检测系统
CN107831217A (zh) 一种填筑连续压实声波检测系统与方法
CN107272534A (zh) 一种智能压实监控系统
Liu et al. Study on real-time construction quality monitoring of storehouse surfaces for RCC dams
CN110331639A (zh) 一种可自主作业智能压路机系统
CN203429539U (zh) 沥青路面施工质量高精度质控系统
Tian et al. Development of real-time visual monitoring system for vibration effects on fresh concrete
CN112051817B (zh) 一种土石坝智能摊铺监控系统及其监控方法
CN106681223A (zh) 一种堆石坝坝面洒水量智能监控系统及其监控方法
Yao et al. Intelligent compaction methods and quality control
Zhan et al. Intelligent paving and compaction technologies for asphalt pavement
CN110197333A (zh) 基于碾压过程实时监控与pda移动检测的面板堆石坝填筑施工全过程管控方法
CN115162309B (zh) 一种接触式碾压机集成压实动量检测系统与方法
Huang et al. Research on real-time supervisory system for compaction quality in face rockfill dam engineering

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 18925075

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 18925075

Country of ref document: EP

Kind code of ref document: A1