Path-Following Control and Terramechanics Analysis for Planetary Rovers Under Wheel-to-Wheel Traction Asymmetry

Ryuya Matsuoka; Keisuke Takehana; Kentaro Uno; Toshinori Kuwahara; Kazuya Yoshida

cs.ROSep 30, 2026

Path-Following Control and Terramechanics Analysis for Planetary Rovers Under Wheel-to-Wheel Traction Asymmetry

Authors: Ryuya Matsuoka, Keisuke Takehana, Kentaro Uno, Toshinori Kuwahara, Kazuya Yoshida

Organizations: Department of Aerospace Engineering, Graduate School of Engineering, Tohoku University, Sendai, Miyagi 980-8579, Japan · Research Center for Green X-Tech, Green Goals Initiative, Tohoku University · Research Center for Space Cross-Tech, Green Goals Initiative, Tohoku University

Abstract

This paper proposes a control strategy for path following that is model-free and relies solely on deceleration for skid-steering planetary rovers navigating deformable loose terrain under continuously imposed traction asymmetry. While conventional controllers that are based on kinematics frequently cause slip-sinkage entrapment by accelerating the wheels during path correction, the proposed approach prevents this failure by setting an upper limit on the maximum commanded velocity. Heading correction is achieved solely through the selective deceleration of the outer wheels, which are located on the outside of the turn, driving them into a negative slip regime to act as a mechanical anchor. The system was evaluated using the four-wheel independent-drive rover EX1 under an asymmetric wheel configuration with different left and right grouser heights that induces significant deviations from the path. Experimental results demonstrate that this deceleration-only control successfully suppresses accumulated lateral drift across various velocity regimes up to 0.7 m/s without causing sinkage. Crucially, direct force measurements from onboard multi-axis sensors provide important empirical evidence of the underlying terramechanics, proving that the targeted deceleration establishes dynamic load equalization across the chassis and completely restores the native thrust capability of the opposite driving wheel.

Figures & tables

Fig. 1: Overview of the experimental rover EX1 used in this study. The rover employs a four-wheel configuration and is equipped with onboard force/torque sensors for measuring wheel-terrain interaction during locomotion on loose soil.

Fig. 2: Asymmetric wheel configuration used to generate a sustained traction imbalance. Different grouser heights (15.0 mm on the left and 5.4 mm on the right) produce unequal wheel-soil interaction forces, resulting in a persistent yaw disturbance.

Fig. 3: Block diagram of the proposed deceleration-only control system for path following. The corrective yaw moment is generated strictly through deceleration by keeping the maximum commanded velocity at or below $v_{\mathrm{ref}}$ .

Parameter	Value
Dimensions (L $\times$ W)	$820\,\mathrm{mm}\times 520\,\mathrm{mm}$
Mass	Approx. $24\,\mathrm{kg}$
Drive system	4-wheel independent drive
Steering mechanism	Skid-steering
Maximum speed	$1.0\,\mathrm{m/s}$
Internal sensors	9-axis IMU, 6-axis force/torque sensors

TABLE I: Specifications of the experimental rover EX1

Fig. 4: Sequential snapshots of the traversal experiment using the EX1 rover on loose soil under the asymmetric configuration.

Fig. 5: Measured trajectories of the rover with the asymmetric wheel configuration on loose soil at target speeds of

0.3

0.5

, and

0.7\,\mathrm{m/s}

. The plotted solid and dashed curves represent the median trajectory filtered from three independent trials.

Fig. 6: Dynamic response of command velocities and traction forces with asymmetric wheels at

0.7\,\mathrm{m/s}

. The black dotted line in (b) denotes the target nominal speed of

0.7\,\mathrm{m/s}

Target Speed ( $v_{\mathrm{ref}}$ )	Wheel Component	Control OFF (Open-Loop)	Control ON (Proposed)	Net Improvement ( $\Delta$ )
	Left Wheel ( $F_{x\mathrm{L}}$ , $15.0\,\mathrm{mm}$ )	$6.00\pm 3.35\,\mathrm{N}$	$-8.22\pm 7.00\,\mathrm{N}$	—
$0.3\,\mathrm{m/s}$	Right Wheel ( $F_{x\mathrm{R}}$ , $5.4\,\mathrm{mm}$ )	$-5.42\pm 3.28\,\mathrm{N}$	$2.71\pm 4.70\,\mathrm{N}$	$\mathbf{+8.13\,\mathrm{N}}$
	Left Wheel ( $F_{x\mathrm{L}}$ , $15.0\,\mathrm{mm}$ )	$7.56\pm 2.83\,\mathrm{N}$	$-10.85\pm 4.60\,\mathrm{N}$	—
$0.5\,\mathrm{m/s}$	Right Wheel ( $F_{x\mathrm{R}}$ , $5.4\,\mathrm{mm}$ )	$-6.04\pm 2.92\,\mathrm{N}$	$3.67\pm 3.82\,\mathrm{N}$	$\mathbf{+9.71\,\mathrm{N}}$
	Left Wheel ( $F_{x\mathrm{L}}$ , $15.0\,\mathrm{mm}$ )	$7.54\pm 2.96\,\mathrm{N}$	$-9.70\pm 4.43\,\mathrm{N}$	—
$0.7\,\mathrm{m/s}$	Right Wheel ( $F_{x\mathrm{R}}$ , $5.4\,\mathrm{mm}$ )	$-5.83\pm 3.79\,\mathrm{N}$	$3.02\pm 3.41\,\mathrm{N}$	$\mathbf{+8.85\,\mathrm{N}}$

TABLE II: Quantitative Summary of Steady-State Traction Performance Based on All Independent Trials

Target Speed	Control	Left Wheel $F_{z}$	Right Wheel $F_{z}$
( $v_{\mathrm{ref}}$ )	State	( $15.0\,\mathrm{mm}$ ) [N]	( $5.4\,\mathrm{mm}$ ) [N]
	OFF	$59.90\pm 26.21$	$75.78\pm 23.81$
$0.3\,\mathrm{m/s}$	ON	$70.20\pm 24.89$	$72.50\pm 27.39$
	OFF	$46.19\pm 5.12$	$64.54\pm 5.17$
$0.5\,\mathrm{m/s}$	ON	$55.13\pm 6.82$	$57.90\pm 4.86$
	OFF	$45.73\pm 6.14$	$65.20\pm 6.87$

TABLE III: Quantitative Comparison of Dynamic Normal Loads Showing Load Equalization under Proposed Control

Panel of Figure 10 — Fig. 7: Quantitative comparison of steady-state mean traction forces under open-loop (OFF) and proposed control (ON) pooled from all independent trials.

Explore similar work

Learning All-Terrain Locomotion for a Planetary Rover with Actively Articulated Suspension
Jun 5, 2026Arthur Bouton, Tristan D. Hasseler, Michael Paton +5Terrain Autonomous Surface Vehicles
Jet Propulsion Laboratory, California Institute of Technology, Pasadena, CA 91109 USA · Center for Autonomy, University of Texas at Austin, Austin, TX 78712 USA · Space Systems Laboratory, University of Maryland, MD 20742 USA
Terrain-Adaptive Grouser Wheel for Optimal Planetary Exploration: Design and Experimental Investigation
May 23, 2026Vincent Griffo, Yashwanth Kumar NakkaAutonomous Surface Vehicles Terrain
Aerospace Robotics Lab, Daniel Guggenheim School of Aerospace Engineering, Georgia Institute of Technology, Atlanta, GA, USA
Parameter-Dependent LMI Synthesis for Semi-Global Differential ISS Trajectory Tracking of Nonholonomic Mobile Robots Under Multiplicative Wheel Slip
Aug 8, 2026Mohammad SabouriSlipsense Wheel
Department of Informatics, Bioengineering, Robotics, and Systems Engineering (DIBRIS), University of Genoa, Via Opera Pia 13, Genoa 16145, Italy · Applied Control & Robotics Research Laboratory (ACRRL), Department of Power and Control Engineering, School of Electrical and Computer Engineering, Shiraz University, Shiraz, Iran

Jun 5, 2026, cs.ROJ/K move · Enter open · S save

Learning All-Terrain Locomotion for a Planetary Rover with Actively Articulated Suspension

Arthur Bouton, Tristan D. Hasseler, Michael Paton, Travis Brown+4

Jet Propulsion Laboratory, California Institute of Technology, Pasadena, CA 91109 USA · Center for Autonomy, University of Texas at Austin, Austin, TX 78712 USA · Space Systems Laboratory, University of Maryland, MD 20742 USA

Terrain Autonomous Surface Vehicles Obstacle Avoidance

Open paper AskPDF arXiv

This paper presents ERNEST, a four-wheeled planetary rover concept equipped with a two-degree-of-freedom Active Gimbal Suspension that combines yaw and roll actuation to enable wheel reconfiguration, steering, and active load redistribution. A single neural network controller, trained to track a desired path across challenging terrain, fully unlocks the capabilities of this actuated suspension system for autonomous obstacle negotiation. A reinforcement learning framework is developed using the high-fidelity DARTS simulation engine, which combines rigid-contact dynamics and Bekker-Wong terramechanics, enabling the emergence of locomotion strategies adapted to loose-soil conditions. To obtain a single unified controller across heterogeneous terrains, a policy consolidation strategy merges the experience of terrain-specialized agents into one neural network, eliminating the need for explicit terrain classification and controller switching. The resulting controller operates on a combination of proprioceptive and exteroceptive feedback, including sparse stereo-derived terrain elevation, chassis attitude, joint states, and force-torque measurements. Zero-shot transfer to the physical rover is achieved through domain randomization, sensor noise injection, and model-to-real system identification. Experimental results demonstrate autonomous traversal of rock fields, a Bickler trap (bump obstacle), a wheel-high step, sand ripples, and sandy slopes. On a 20° sandy slope, the learned controller reduces the cost of transport by 37% on dry sand despite the additional actuation, and achieves superior performance on wet sand where the passive suspension becomes completely immobilized. A video accompanying this paper is available at https://youtu.be/d684P5a3xMc

Abstract

Explore similar work

Learning All-Terrain Locomotion for a Planetary Rover with Actively Articulated Suspension

Terrain-Adaptive Grouser Wheel for Optimal Planetary Exploration: Design and Experimental Investigation

Parameter-Dependent LMI Synthesis for Semi-Global Differential ISS Trajectory Tracking of Nonholonomic Mobile Robots Under Multiplicative Wheel Slip