Wheel Loader Research Articles

Optimizing bucket-filling strategies for wheel loaders inside a dream environment

Reinforcement Learning (RL) requires many interactions with the environment to converge to an optimal strategy, which makes it unfeasible to apply to wheel loaders and the bucket filling problem without using simulators. However, it is difficult to model the pile dynamics in the simulator because of unknown parameters, which results in poor transferability from the simulation to the real environment. Instead, this paper uses world models, serving as a fast surrogate simulator, creating a dream environment where a reinforcement learning (RL) agent explores and optimizes its bucket-filling behavior. The trained agent is then deployed on a full-size wheel loader without modifications, demonstrating its ability to outperform the previous benchmark controller, which was synthesized using imitation learning. Additionally, the same performance was observed as that of a controller pre-trained with imitation learning and optimized on the test pile using RL.

Design, Modeling and Control of a Hardware-in-the-Loop Testbed for Off-Road Vehicles

Abstract This paper presents the design, modeling, and control of a hardware-in-the-loop (HIL) testbed for off-road vehicles. The proposed HIL testbed employs a transient hydrostatic dynamometer to load a diesel engine to emulate any loading cycles of a wheel loader, which is a representative off-road vehicle. A fully validated wheel loader model is used to calculate the engine load, including both the drive and work functions. Besides, Iterative Learning Control (ILC) has been designed for the loading torque tracking of the hydrostatic dynamometer to ensure accurate emulation of real-world operation scenarios. The developed HIL testbed is used to demonstrate more than 26% energy benefits of automated wheel loaders through systematic optimization compared with human-operated wheel loaders. This HIL testbed serves as a robust platform for advancing research and development across various off-road vehicles, including excavators, tractors, and harvesters.

A parallel coupling framework for DEM-MBD: Model verification and application

Bulk material handling equipment can be numerically studied using the DEM-MBD method, where the particulate dynamics are solved by DEM (Discrete Element Method) and the motion characteristics of the mechanical components are modeled by MBD (Multi-Body Dynamics). In this work, a two-way coupling framework is proposed for the co-simulation of our self-developed DEM solver (DEMSLab) with the commercial MBD software (RecurDyn) using a parallel staggered coupling approach. To verify the proposed coupling method, the simulation result of a torsional spring damper system is compared with the corresponding analytical solution. After that, to assess the feasibility of this method in industrial-scale applications, a series of simulations involving a wheel loader is carried out. Different simulation conditions are adopted to reveal their impact on computational efficiency and simulation accuracy. These conditions include the mesh representation of the equipment geometry, the MBD time step size, and the coupling interval. An improved mesh simplification algorithm is proposed and used to reasonably simplify the surface mesh of the wheel loader, thus lowering the computational intensity of particle-wall interactions in DEM. The comparison results indicate that the computational time can be largely saved while the simulation accuracy remains considerably high. The MBD step size and coupling interval are crucial for numerical stability and should be selected carefully.

Wheel loader seat damping control research based on ADRC

This paper proposes a new type of damped multi-mode switching damper based on a high-speed switching solenoid valve. The structural design of the damper is completed, and its working principle of realizing different damping modes is analyzed. A damping characteristic model of the damper is constructed, and the change rule of the damping characteristic of the multi-mode switching damper is analyzed. The seat suspension of a wheel loader is equipped with a damper that is controlled by a damping strategy. A 10-degree-of-freedom mixed-logic dynamic model of the seat suspension system is established using HYSDEL, and a multi-mode switching damper is employed. The weight coefficients are adjusted using a self-immunity control strategy, and the hybrid model predictive control method is used to complete the damping control strategy. Simulation results show that, compared with the fuzzy control strategy, the designed wheel loader seat suspension damping active disturbance rejection control (ADRC) strategy not only effectively reduces the root-mean-square (RMS) value of the driver’s vertical vibration acceleration up to more than 40% at different speeds and shock conditions, but also effectively reduces the driver’s vertical vibration acceleration. The ADRC strategy is a robust approach that achieves the desired control objective of the system.

Experimental study of dual-cycle thermal management system for engineering radiator

With the increasing demand for heat dissipation of engineering vehicles, a dual-cycle cooling system is introduced in this paper to prevent the adverse effects of engineering vehicles’ equipment when operating at the overheating temperature. The performance of the new system is analyzed through tests, and the results show that the dual-cycle cooling system can meet the thermal balance requirements of the engineering vehicle during the shovel operation. Compared with the traditional cooling system, the new cooling system improved performance in terms of volume, engine energy consumption and working oil efficiency. The oil consumption of a wheel loader using the dual-cycle cooling system is reduced by 1% per hour, and the temperature of its transmission oil and hydraulic oil is reduced by more than 10 °C. The new cooling system has bright future in energy saving and emission reduction of engineering vehicles.

Data augmentation aided excavator activity recognition using deep convolutional conditional generative adversarial networks

Excavator activity recognition is a crucial task with significant practical applications such as assessing production efficiency, injury prevention, and enabling intelligent control. However, implementing deep learning-based activity recognition methods presents considerable challenges due to the requirement of large training datasets. And sensor-based activity recognition demands more convenient and reliable sensors. This paper proposes a generalized data augmentation framework for excavator activity recognition. Main pump pressure and flow signals are generated using Deep Convolutional Conditional Generative Adversarial Networks (DC-CGAN). The proposed framework integrates Long Short-Term Memory (LSTM) and Convolutional Neural Network (CNN) architectures for activity recognition. Results indicate that the framework outperforms traditional data augmentation methods regarding the quality of generated data and shallow and single deep networks regarding model accuracy and generalization. Additionally, the framework was implemented on three sizes of excavators and a wheel loader, demonstrating excellent versatility.

Dynamic power matching for wheel loader based on power reflux hydrodynamic transmission system

Dynamic power matching for wheel loader based on power reflux hydrodynamic transmission system

Examining the simulation-to-reality gap of a wheel loader digging in deformable terrain

Examining the simulation-to-reality gap of a wheel loader digging in deformable terrain

World Modeling for Autonomous Wheel Loaders

This paper presents a method for learning world models for wheel loaders performing automatic loading actions on a pile of soil. Data-driven models were learned to output the resulting pile state, loaded mass, time, and work for a single loading cycle given inputs that include a heightmap of the initial pile shape and action parameters for an automatic bucket-filling controller. Long-horizon planning of sequential loading in a dynamically changing environment is thus enabled as repeated model inference. The models, consisting of deep neural networks, were trained on data from a 3D multibody dynamics simulation of over 10,000 random loading actions in gravel piles of different shapes. The accuracy and inference time for predicting the loading performance and the resulting pile state were, on average, 95% in 1.2 ms and 97% in 4.5 ms, respectively. Long-horizon predictions were found feasible over 40 sequential loading actions.

Structural load estimation of the wheel loader for customer usage profile monitoring

Structural load estimation of the wheel loader for customer usage profile monitoring

Research on Multi-Mode Variable Parameter Intelligent Shift Control Method of Loader Based on RBF Network

The loader is one of the most widely used pieces of engineering machinery in the world for soil transportation, loading and unloading materials, and low-intensity shovel digging operations in harsh and complex operating conditions; it requires very frequent shifting and has other challenging characteristics. In order to realize automatic frequent shifting, we need to better design the shifting rules in the shifting process, improve the shifting quality and working efficiency, and solve the key engineering problems of energy saving and high efficiency in the shifting process of loaders. In this paper, a 7-ton wheel loader is taken as the research object, the loader shoveling process of the four operating modes is analyzed, and a multi-mode variable parameter shift law is designed. Aiming at the complicated and nonlinear characteristics of the power transmission system of the loader, an intelligent shift control method based on an RBF neural network is proposed. Finally, the simulation test and the clutch shift oil pressure test are carried out. From the test results, the clutch test oil pressure curve obviously shows a four-stage upward trend during shifting, and the buffering effect is obvious. The designed multi-mode variable-parameter intelligent shift law of the loader is reasonable and feasible, and the shift recognition rate reaches 97.92%, which provides theoretical support for the realization of intelligent automatic speed change control of the loader, and it certainly has engineering value.

Synchronized path planning and tracking for front and rear axles in articulated wheel loaders

Wheel loaders are multi-function construction equipment, yet their complex kinematics and coupled front-rear axle relationships pose significant challenges in path planning and tracking control. This paper presents a synchronous path planning method and a reinforcement learning-based tracking controller for these axles. Initially, a kinematic model and tracking error model are developed based on steering radius analysis and verified through simulations. The Reeds-Shepp curve is used to plan the V-shaped path for the front axle, with the rear axle path derived from the kinematic model. Using reinforcement learning, a path tracking controller is developed with a preview model of tracking error. Performance is validated through simulations and field tests, showing a position error reduction of 25% and 37.5% for two paths, and heading error reductions of 27.8% and 27.3%. The results also indicate improved stability in heading and articulation angles during tracking.

System configuration, control development, and in-field validation of a hybrid electric wheel loader featuring electrically-boosted engine

This paper proposes an innovative next-generation wheel loader derived from a production series electric wheel loader by John Deere. The proposed wheel loader features a series hybrid electric powertrain with an energy storage system and an electrically-boosted turbocharged diesel engine. A detailed system design inclusive of powertrain configuration and control development including a novel heuristics-based vehicle power management is presented. A simulation model was created to evaluate the potential fuel savings of the proposed wheel loader, revealing a fuel saving potential of over 10% when compared to the baseline configuration. Subsequent in-field testing of an actual demo wheel loader verified its ability to achieve over 10% fuel savings, thereby confirming the simulation outcomes, demonstrating the promise of the proposed hybrid powertrain, and validating the efficacy of the control system developed in this research.

Fuel Efficiency Analysis and Control of a Series Electric Hybrid Compact Wheel Loader

<div>The escalating demand for more efficient and sustainable working machines has pushed manufacturers toward adopting electric hybrid technology. Electric powertrains promise significant fuel savings, which are highly dependent on the nature of the duty cycle of the machine. In this study, experimental data measured from a wheel loader in a short-loading Y-cycle is used to exercise a developed mathematical model of a series electric hybrid wheel loader. The efficiency and energy consumption of the studied architecture are analyzed and compared to the consumption of the measured conventional machine that uses a diesel engine and a hydrostatic transmission. The results show at least 30% reduction in fuel consumption by using the proposed series electric hybrid powertrain, the diesel engine rotational speed is steady, and the transient loads are mitigated by the electric powertrain. The model also shows that 20% of drive energy could be regenerated through braking using the drive electric motors. Opportunities for engine downsizing are established and the losses are analyzed and compared for both machines. The results show 42% savings in the overall system losses especially the diesel engine and drivetrain.</div>

Bridging simulation granularity in system-of-systems: conjunct application of discrete element method and discrete event simulations in construction equipment design

AbstractThe paper addresses a critical challenge in System-of-Systems (SoS) simulations arising from the different granularity levels in SoS simulations, integrating non-coupled Discrete Element Method results into SoS-level Discrete Event Simulations using surrogate modeling. Illustrated with a wheel loader bucket use-case in mining, it enhances early design decision-making and lays the groundwork for improving SoS simulations in construction equipment design. This paves the way for broader research and application across diverse engineering design domains.

Bayesian Optimization for Digging Control of Wheel-Loader Using Robot Manipulator

Wheel loaders are construction machines that are mainly used for excavating and loading sedimented ground into dump trucks. The objects to be excavated range from large materials, such as blast rock and crushed stone, to small materials, such as gravel, slag, and coal ash. Therefore, the excavation operation of wheel loaders requires a high skill level to cope with various terrains and soil types. As worker numbers at quarry sites decline, developing highly automated technology to replace operators is crucial. In particular, the geometry of the ground to be excavated by the wheel loader changes with each excavation, so the control parameters must be adapted sequentially during automated excavation. In this study, we proposed an online learning method using Bayesian optimization to search for control parameters from multiple trials and modify them sequentially. In particular, we formulate a multi-objective optimization problem maximizing a weighted linear combination of the payload and workload as an objective function. To validate the proposed method, we constructed an environment in which repeated digging tests can be performed using a robot manipulator with a bucket attached. Through comparative tests between feed-forward control, in which the robot moves along a fixed trajectory independent of the digging reaction force, and off-line control, in which the robot modifies the digging trajectory in response to the digging reaction force, we compared the ability of these methods to cope with terrain volume that is different from that of the optimization trial.

Testing the suitability of vector normalization procedure in topsis method: application to wheel loader selection

The object of the research consists of testing the suitability of the vector normalization procedure (NP) in the Technique for Order Preference by Similarity to the Ideal Solution (TOPSIS) method. One of the most problematic steps of the Multi-Criteria Decision Making (MCDM) process is related to the application of NPs by default to transform different measurement units of criteria into a comparable unit. This is because of the absence of a universal agreement that defines which NP is the most suitable for a given MCDM method. In the literature, there are thirty-one available NPs, each one of them has its strengths and weaknesses and, accordingly, can efficiently be applied to an MCDM method and even worst to another. Let’s note that many NPs (e. g., NPs of sum, max-min, vector, and max) have been used by default (i. e., without suitability study) in the TOPSIS method. Consequently, outcomes of multi-criteria evaluation and rankings of alternatives considered in the decision problems could have led to inconsistent solutions, and, therefore, decision-makers could have made irrational or inappropriate decisions. That’s why suitability studies of NPs become indispensable. Moreover, a description of the methodology, proposed in this research, is outlined as follows: 1) method of weighting based on an ordinal ranking of criteria and Lagrange multiplier (for determining criteria weights); 2) TOPSIS method (for ranking considered alternatives); 3) a statistical approach with 3-estimate (for comparing effects generated by the used NPs). In the research, twelve different NPs are compared to each other in the TOPSIS method via a numerical example, which deals with the wheel loader selection problem. The results of the comparison indicate that, amongst the twelve different NPs analyzed in this suitability study, vector NP has the lesser effect on the considered alternatives’ evaluation outcomes, when used with the TOPSIS method. The vector NP-TOPSIS approach can therefore be applied to solve multi-criteria decision problems. Its application further allows the decision-makers and users to better select efficient solutions and, consequently, to make conclusive decisions.

Design and Control Performance Optimization of Variable Structure Hydrostatic Drive Systems for Wheel Loaders

The traditional loader drive system is based on the hydraulic torque converter as the key component, and its gear is shifted through the coordination of the clutch and gearbox, which greatly increases power loss and operator fatigue. To address the above problem, a variable-structure hydrostatic drive system is proposed. This system adopts a closed-loop design and adjusts the displacement of the pump and dual motors to follow the throttle opening of the vehicle, achieving automatic gear shifting and smooth speed regulation. It can also automatically change the system structure according to the vehicle’s speed, meeting the vehicle’s demand for rapid switching output of high torque and high speed. At the same time, in the displacement matching control process, an adaptive sliding mode control scheme based on radial basis function neural network compensation is proposed. This scheme designs an adaptive hyperparameter update strategy according to the characteristics of the system to effectively compensate for changes in uncertain factors. Experimental results show that, compared to traditional drive systems, this system has the characteristics of simple operation, smooth speed regulation, and high fuel economy.

PMF-SLAM: Pose-Guided and Multiscale Feature Interaction-Based Semantic SLAM for Autonomous Wheel Loader

PMF-SLAM: Pose-Guided and Multiscale Feature Interaction-Based Semantic SLAM for Autonomous Wheel Loader

Energy-Saving Impact and Optimized Control Scheme of Vertical Load on Distributed Electric Wheel Loader

During the operation of a wheel loader, the external load acting on the bucket undergoes many changes, resulting in significant changes in the load ratio on the front and rear axles. For this reason, controlling a standard wheel loader is not trivial. In addition, in the case of a distributed electric wheel loader (DEWL), the operating control algorithm is often complex and is, therefore, the subject of optimization studies. This study compared the electric power consumption across different vertical loads, speeds, and travel directions for single-front, single-rear, and dual-motor configurations, both during transporting and pre-shoveling operations. The analysis led to the development of control rules based on energy-saving objectives. Under the shoveling condition, it was observed that vertical loads can lead to an insufficient driving force and skidding, necessitating the proposal of a new optimized control scheme. The results revealed that the optimal solution for transporting is the single-motor drive control scheme without a mechanical connection between the front and rear motor. With the single-motor control scheme, comparing the preferred controlled motor with the unselected motor under different loads, the average electrical power savings for forward, backward, and circling were at least 3.51%, 3.12%, and 0.34%, respectively. Under the pre-shoveling condition, the optimal control scheme was identified as the single rear motor control scheme, effectively reducing electrical power consumption. In response to the issues encountered during the shoveling condition, an economical solution involving the modification of the front axle transmission ratio has been proposed, along with an optimized control scheme based on vertical load variations.