No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Demonstration of Snow Removal Work by Wheel Loader in an Environment Surrounded by Obstacles
Abstract
Snow removal work using construction equipment faces problems such as a shortage of skilled operators owing to the declining birthrate and aging population, work in dangerous areas, and accidents caused by a lack of concentration during long work hours. To improve the working environment, research and development of automation of construction equipment are actively conducted. Therefore, in this study, we aim to generate a driving path for wheel loaders for snow removal work in a work environment surrounded by obstacles, such as walls and fences. Furthermore, the proposed method considers the changing shape of the snow piles during the removal. We experimentally verified that snow removal could be performed using an actual wheel loader on the route generated by the proposed simulation.
... Given the limitations of this approximate method, collecting point cloud data of the material pile using LiDAR is a viable option for achieving 3D reconstruction. Reference [4] has installed multiple LiDAR in the working environment to achieve distance measurement and 3D reconstruction of the pile. This solution allows the loader to obtain the shape changes in the pile in real time during the shoveling operation. ...
... Finally, the curve is fitted, and the average curvature value is calculated. The formula is shown in (4). Assume that the equation for the curve is shown in (3). ...
Gravel is the most frequently used material in infrastructure construction. However, the irregular shape of the gravel pile makes it challenging for the loader to predict a stable shoveling position, which can easily result in partial collapse or a complete landslide, thereby posing a serious threat to the equipment. In view of the imperfect method of determining the shoveling position of the pile by the current unmanned loader and the high hardware requirements for the deployment of the identification model, this paper first establishes a mathematical model of the loader, and preliminarily determines the influence of the concave and convex edges of the gravel pile on the shoveling position selection through discrete element joint simulation; secondly, the influence of the pile with different edge curvatures on the loader operation process is analyzed in the simulation software, and the radar map is used to further identify the superior position features; finally, Ghost Net is used as the backbone network, the RFB module is introduced into the Backbone, and the CBAM attention mechanism is integrated into the C3 module to identify the lightweight YOLOv5s shoveling position. Discrete element analysis and a lightweight network model were used in the above study to find the safest and most effective shoveling positions. During the test that mimicked how the loader would actually shovel, the number of parameters in the improved model was cut down to 32.5% of the original, the number of calculations was cut down to about 55.2% of the original, and the average accuracy of finding the shoveling position of the gravel pile reached 98%.
In this study, we propose an automated excavation system for pneumatic caisson construction using either a single excavator or multiple excavators. The system divides the work area among the excavators and manages the transition of occupied states in shared areas to reduce the risk of collisions. Additionally, the Laplace potential method is employed for path planning to avoid collisions with equipment inside the caisson. The system also includes a mechanism for disposing of soil outside the caisson by dumping it into an earth bucket. To confirm the effectiveness of the proposed method, verification tests were conducted: one using a single excavator in a narrower-than-usual caisson, and another using two excavators in a test field. These tests demonstrated the method’s effectiveness.
Ensuring the transportation process in the Siberian region during the winter months involves protecting the railway infrastructure from external climatic factors, such as snow drifts and icing. To ensure timely clearing of tracks, a wide fleet of snow removal equipment on rails is used, the most numerous of which is the snow removal train type SM-2 and its modifications. The technological cycle of such trains can be reduced to three operations: clearing the track and loading snow mass, transportation to the unloading dead end, unloading and returning to the work site. Reducing both fuel and time resources in each technological link is an urgent task for railways. The purpose of the study is to evaluate the capabilities of monitoring the parameters of a diesel generator for more effective planning of operating modes of snow removal equipment. The current strength of three phases of a diesel generator was measured in various operating modes. The values of the current strength when starting electric drives of various train powers are estimated. The duration of operation of a diesel generator in nominal mode is compared with the operational need to set the operating mode in the technological cycle of clearing tracks from snow.
This study introduces expedite the complete transfer of distributed gravel piles with an automated wheel loader. The wheel loader scoops the gravel and unloads it onto the bed of a truck. The total mileage for the repeated scooping and unloading work is reduced. The complicated optimal task is divided into three simple, organized subjects and different appropriate algorithms are used for each subject. A method is proposed that interactively selects the appropriate scooping points, unloading points, and appropriate path for the varying shape of the gravel pile until all gravel piles have been scooped up. To expedite the complete transfer of distributed gravel piles, the deep reinforcement learning model, trained via simulation, can be used in practice by implementing as part of the proposed methods. This paper describes these issues and the proposed methods. The performance of the actual application is demonstrated in terms of the calculation time and the feasibility of the simulations.
This paper presents two autonomous pile transfer experiments with a robotic compact wheel loader. The main objectives were to achieve automated load-haul-dump work cycles which successfully transported material, and also updated the scooping and dumping location if needed based on the task-level plan and changing worksite topography. This would then demonstrate the potential for the system to work autonomously until job completion. A task-level plan consisting of a scooping and dumping area is first specified using a 3D graphical user interface. This consists of interactive graphical tools rendered on a surface model of the site, which is obtained using an on-board laser scanner. Commands for loading, hauling and dumping actions are then generated and followed autonomously without any further human input. The experiments include a 5-cycle test over 9 minutes, though with 2 unsuccessful scooping actions, and a 10-cycle test over 24 minutes, with one unsuccessful scooping action. During these tests, new ground models were made during each cycle to assess the current topography, which resulted in updated scooping and dumping locations. Several areas for improvement were identified, such as implementing a more effective loading strategy and more efficient path planning. Overall the system shows potential for enabling even longer-duration autonomous operations, and to be used for comparing different planning and control strategies.
Automation of earth-moving industries (construction, mining and quarry) require automatic bucket-filling algorithms for efficient operation of front-end loaders. Autonomous bucket-filling is an open problem since three decades due to difficulties in developing useful earth models (soil, gravel and rock) for automatic control. Operators make use of vision, sound and vestibular feedback to perform the bucket-filling operation with high productivity and fuel efficiency. In this paper, field experiments with a small time-delayed neural network (TDNN) implemented in the bucket control-loop of a Volvo L180H front-end loader filling medium coarse gravel are presented. The total delay time parameter of the TDNN is found to be an important hyperparameter due to the variable delay present in the hydraulics of the wheel-loader. The TDNN network successfully performs the bucket-filling operation after an initial period (100 examples) of imitation learning from an expert operator. The demonstrated solution show only 26% longer bucket-filling time, an improvement over manual tele-operation performance.
The paper is about finding the global optimum for a wheel loader work cycle in a gravel application. This includes simulating the gravel and extracting the trajectories for the main actuators; propulsion, lift and tilt, during the work cycle. The optimal control method is dynamic programming and the optimum is calculated with regard to fuel efficiency [ton/l] but can be weighted towards productivity [ton/h].
The analytical optimal control results are compared to an extensive empirical measurement done on a wheel loader and shows around 15% higher fuel efficiency compared to the highest fuel efficiency measured among real operators.
In construction projects, wheel loader is widely used for transporting building materials. The operations of wheel loader contain extensively repetitive tasks that could be optimized and automated for sustainable construction. This paper proposes, according to optimal control theory, an optimization approach for the loading cycle concerning fuel efficiency and environmental impacts. Based on a 2D space discretization and a detailed wheel loader model, a dynamic programming approach is formulated for optimal path search. The method is effective for integrating the system models with the path planning algorithm and for handling objective functions and constraints. A case study was carried out to demonstrate the application of the proposed algorithm for the loading cycle with spatial constraints.
Wheel loaders in the V-type loading cycle are characterized by complicated loading conditions, nonlinear power-train system, and time-variable engine power distribution. Therefore, it is difficult to predict the fuel consumption of wheel loaders in the V-type loading cycle. The static matching methods cannot provide fuel consumption prediction for the loading cycle. In this paper, the prediction method and model of the fuel consumption for wheel loaders in the V-type loading cycle were proposed. Firstly, the hydraulic system data were tested when a wheel loader loaded three different materials in a typical V-type loading cycle. Secondly, the tested data were filtered by the 8th-order Butterworth filter and the dimensionless power deduction equations of hydraulic power system for loading three different materials were obtained by Gaussian and linear fitting based on the filtered data in the loading cycle. Finally, fuel consumption was obtained with the compiling dynamic calculation program as well as input parameters of tested vehicle speed, throttle parameter, and the dimensionless equation. The simulation results agreed well with experiment results. Dynamic calculation program is applicable to calculate loading economy and can provide academic guidance for wheel loader’s design and optimization.
Path planning is an essential task for the navigation and motion control of autonomous robot manipulators. This NP-complete problem is difficult to solve, especially in a dynamic environment where the optimal path needs to be rerouted in real-time when a new obstacle appears. The ACO (Ant Colony Optimization) algorithm is an optimization technique based on swarm intelligence. This paper investigates the application of ACO to robot path planning in a dynamic environment. Two different pheromone re-initialization schemes are compared and computer simulation results are presented.
| Most path planners for car-like robots compute Reeds and Shepp paths" made up of line segments connected with circular arcs. Such paths have a discontinuous curvature that makes them dicult to track (curvature is related to the orientation of the front wheels). The purpose of this paper is to present one of the rst path planner for car-like robots that computes paths with continuous-curvature and upper-bounded curvature derivative (curvature derivative is related to the steering velocity). The approach presented herein relies upon a steering method, i.e. an algorithm that computes paths without taking into account the obstacles of the environment, which is then embedded within a general path planning scheme in order to deal with the obstacles and thus solve the full problem. The paths computed are made up of line segments, circular arcs and clothoid arcs. Keywords | mobile-robot, path-planning, non-holonomic-system Acknowledgements | this work was partially supported b...
Localization methods for autonomous construction vehicles include real‐time kinematic positioning through a global navigation satellite system (GNSS) and a scan matching with a light detection and ranging (LiDAR) attached to mobility. However, these conventional methods have low estimation accuracy when the vehicle's surroundings have few features and the vehicle is in the no‐GNSS area. For the estimation in such areas, this paper proposes a localization method that can estimate by matching a 3D model of a construction vehicle with a point cloud obtained from 3D LiDARs installed in a work area. To realize the high‐accuracy and high‐speed processing of the localization, we propose remodeling using the predictive motion model (RM) algorithm to modify the 3D model in the registration process. In the experimental results on rough terrain, we confirmed that our method can estimate a vehicle's position and yaw angle with accuracies of 0.121 m and 0.016 rad, respectively. In addition, compared with the case without the RM algorithm, the construction vehicle's position and yaw angle accuracies improved up to 5 and 12 times, respectively.
We aim to work automation of gravel pile transportation by construction machinery. To automate civil engineering work with construction machinery, it is necessary to control by remote control. Retrofit kits are used to remotely control and automate traditional construction machinery. Retrofit kits are sold by domestic manufacturers. To autonomously control a retrofitted construction machine, it is necessary to make a detailed model, which is difficult. In this paper, we retrofitted the wheel loader and implemented a remote control system. we researched the control characteristics by collecting data on the retrofit kit mechanism and vehicle motion when controlling a retrofitted wheel loader.
Authors have been engaged in the research on the outdoor working robotics, and targeting the autonomous pile loading operation by a wheel loader in mining particularly. Authors developed a real sized autonomous wheel loader "Yamazumi-4," and succeeded the unmanned autonomous pile loading operation experimentally. The precise traveling control is necessary to repeat loading operations. The real sized construction machines like the wheel loader are difficult to control the travelling speed precisely. Authors adopted the path planning and path following control based on the odometer and succeeded to gain precise geometric shape of the path. This paper describes the path planning and path flowing control of the autonomous wheel loader.
The authors have developed the autonomous loading system by wheel loader. In this paper, we propose the path planning algorithm to optimally place both one unloading point and a set of several scooping points corresponding to that point. For simultaneously determining the set and streamlining the whole series of operations, this algorithm include the step of estimating the shape of rock piles collapsing by scooping works.
We propose a method for the efficient scooping of sediment by a wheel loader. We performed experiments to verify the effectiveness of the proposed method. First, in order to calculate the energy required for the scooping operation, we derived a model of the reaction force that is generated on the bucket. We verified the validity of the model by experiments. We then used the reaction force model to determine the optimal trajectory for the bucket. A mechanical simulator was used to verify the trajectory. Although experimental results showed the unexpected effects of sediment that settled inside the bucket, we obtained results that reflect theoretical values.
A wheel loader is the major vehicle that has been used for scooping, carrying and loading of crushed ores in open pit mines. This paper describes a path generation method for autonomous control of the wheel loader. From the viewpoint of stability of movement and easy control to move along the designed path, smoothness and simple calculation of command value are necessary for the path. In order to satisfy these conditions, a path generation method using a clothoid curve is proposed. A clothoid curve has a continuous changing curvature that is proportional to the length of the curve, and provides continuous changing of steering angle. Furthermore, the interpolation between the start and end point of the loader is proposed, which consists of clothoid curve segment and straight line segment. V-shape curve including backward path and forward path is also generated with considering the avoidance of rapid turning. INTRODUCTION A wheel loader is the major vehicle that has been used for scooping, carrying and loading of crushed ores in open pit mines. Usually, this machine is operated by an expert operator. But, if this machine is used in the dangerous field such as the disaster place or in the task to develop the space resources, the unmanned automation technology is necessary. In order to realize the full automation of wheel loader task, the loader needs to have an ability to recognize the working environment, to plan the scooping and loading task and to execute the task autonomously. Various researches have been already conducted for mobile robots. However, since the working environment of a wheel loader is rough terrain and is changing according to the advance of the task, the full automation of wheel loader task is much more complicated as compared to another types of mobile robots that are operated in the artificial environment. Furthermore, an articulate steering is used for the wheel loader, which is peculiar to a wheel loader. Although the articulate steering has an advantage of small steering radius, the trajectory tracking control is very complicated because the tire is easy to slip in steering. The purpose of this study is to investigate the automation technology to achieve the full automation of the wheel loader task. In this paper, only path generation is described.
There is a demand of mining automation by heavy equipment in an open mine. Authors aim to automization of motion for wheel loader in scooping and loading operation. In this paper, we applied the optimization method by Genetic Algorithm to the path planning for the wheel loader on scooping and loading operation. We showed the quasi-optimized path and demonstrated by the miniature wheel loader robot.
This paper describes a new scan matching method for mobile robot localization and mapping. The proposed method "intensity-ICP" uses laser reflection intensity obtained by a laser range scanner named "SOKUIKI sensor". Compared with conventional scan matching methods which are effective just in geometric featured environments, intensity-ICP is effective in both geometric featured and featureless environments if there are some features like colors or materials. This method can build a map with laser reflection intensity data. The map will be effective to robust localization because it has abundant information; not only geometric data but also laser reflection intensity
The authors have been conducting research on an autonomous system for loading operation by wheel loader. Experimental results at a field test site using full-size model (length: 6.1m) will be described in this paper. Basic structure of system consists of three sub systems: measuring and modeling of environment, task planning and motion control. The experimental operation includes four cycles of scooping and loading to dump truck. The experimental results prove that the developed system performed the autonomous operation smoothly and completed the mission.