cs.ROAug 8, 2025

All-Electric Heavy-Duty Robotic Manipulator: Surrogate-Enhanced Modeling and Adaptive Modular Control

Authors: Amir Hossein Barjini, Mohammad Bahari, Mahdi Hejrati, Jouni Mattila

Organizations: Department of Engineering and Natural Sciences, Tampere University, 7320 Tampere, Finland

Read PDF
Ask AI
Cite

Copy citation

DOINot available for this paper

Abstract

This paper presents a unified system-level modeling and control framework for an all-electric heavy-duty robotic manipulator (HDRM) driven by electromechanical linear actuators (EMLAs). A surrogate-enhanced actuator model, combining integrated electromechanical dynamics with a neural network trained on a dedicated testbed, is integrated into an extended virtual decomposition control (VDC) architecture augmented by a natural adaptation law. The derived analytical HDRM model supports a hierarchical control structure that seamlessly maps high-level force and velocity objectives to real-time actuator commands, accompanied by a Lyapunov-based stability proof. In multi-domain simulations of both cubic and a custom planar triangular trajectory, the proposed adaptive modular controller achieves sub-centimeter Cartesian tracking accuracy. Experimental validation of the same 1-DoF platform under realistic load emulation confirms the efficacy of the proposed control strategy. These findings demonstrate that a surrogate-enhanced EMLA model embedded in the VDC approach can enable modular, real-time control of an all-electric HDRM, supporting its deployment in next-generation mobile working machines.

Figures & tables

Fig. 1: Schematic of the electric HDRM.
Fig. 2: VDC frames of the manipulator.
Fig. 3: Schematic of the lift (3-bar mechanism).
Fig. 4: Schematic of the tilt (4-bar mechanism).
Fig. 5: Comparison of the experimental force and velocity responses with the desired trajectory.
Fig. 6: Surrogate-enhanced efficiency map of the EMLA, combining physics-based modeling with DNN predictions. The map illustrates actuator efficiency across varying load forces and linear velocities, capturing both model-driven dynamics and experimentally-informed nonlinearities.
Fig. 7: Schematic view of the trajectories. a) Cubic trajectory with 2020 cm length. b) Planar triangular trajectory.
Fig. 8: Performance comparison of controllers using the cubic trajectory.
Fig. 9: Performance comparison of controllers using the planar triangular trajectory.
UncertaintyCubic TrajectoryTriangular Trajectory
+40 %0.0024 m0.0042 m
+20 %0.0019 m0.0035 m
0 % (Nominal)0.0015 m0.0029 m
-20 %0.0029 m0.0050 m
-40 %0.0036 m0.0060 m
TABLE I: RMSE of the proposed controller under parametric uncertainties for two trajectories.
Fig. 10: Experimental setup used to validate the proposed control strategy. The testbed includes the Nidec PMSM, a reduction gearbox, and a high-precision screw forming the EMLA, as well as electrohydraulic cylinder as the load emulator.
Fig. 11: Experimental tracking results for the lift actuator under the cubic trajectory, illustrating tracking accuracy under three-dimensional motion.
Fig. 12: Experimental tracking results for the tilt actuator under the cubic trajectory, illustrating tracking accuracy under 3D motion.
Fig. 13: Experimental tracking results for the lift actuator under the planar triangular trajectory, illustrating performance under two-dimensional constrained motion.
Fig. 14: Experimental tracking results for the tilt actuator with planar triangular trajectory, illustrating performance under 2D constrained motion.
More in this topic

Robotic Control

Robotic control studies how machines use sensing and computation to produce reliable movement and interaction in changing environments.

See all 583 papers

Explore similar work

CardsList

  1. Learning-Based Dynamics Modeling and Robust Control for Tendon-Driven Continuum Robots

    Apr 28, 2026Ziqing Zou, Ke Qiu, Fei Wang +3Gated Recurrent UnitRobot Trajectory Tracking

    Department of Control Science and Engineering, Zhejiang University, Hangzhou, China

  2. Precision at Speed: Sample-Efficient Online Model-Based Reinforcement Learning for Hydraulic Excavator Control

    Sep 25, 2026Claudio Canales, Fang Nan, Marco Hutter +1Robotic ControlModel-Based RL

    Robotic Systems Lab, ETH Zürich, 8092 Zürich, Switzerland · Advanced Mining Technology Center (AMTC) and the Department of Electrical Engineering, Universidad de Chile, Santiago, Chile

  3. NeuralActuator: Neural Actuation Modeling for Robot Dynamics and External Force Perception

    Jul 13, 2026Zhiyang Dou, John U. Onyemelukwe, Hangxing Zhang +9Neural Surrogate ModelingRobot Dynamics Identification

    1MIT · 2Amazon Robotics