Failure Mechanisms and Risk Estimation for Legged Robot Locomotion on Granular Slopes
Xingjue Liao and Feifei Qian · 2026
Read Failure Mechanisms and Risk Estimation for Legged Robot Locomotion on Granular Slopes by Xingjue Liao and Feifei Qian (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2603-06928.txt.
Introduction
Liao and Qian study how legged robots fail on granular slopes like dunes, where media lose shear strength and yield under gravity. With a six-legged robot on a tiltable bed, they measure speed plus normal and shear resistive forces. Normal penetration resistance barely changes with tilt, while shear resistance falls as slope angle grows. Their robot-terrain model predicts anchoring, step length, and speed from terrain strength and slope angle, and maps failure diagrams. Slope losses come from delayed anchoring and backward slip, not sinkage.
Catalog record
| CREATOR | Xingjue Liao and Feifei Qian |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 33600 |
| LICENSE | http://creativecommons.org/licenses/by-sa/4.0/ |
| RIGHTS | CC BY SA |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2603.06928
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Failure Mechanisms and Risk Estimation for Legged Robot Locomotion on Granular Slopes Xingjue Liao and Feifei Qian Abstract Locomotion on granular slopes such as sand dunes remains a fundamental challenge for legged robots due to reduced shear strength and gravity-induced anisotropic yielding of granular media. Using a hexapedal robot on a tiltable granular bed, we systematically measure locomotion speed together with slope-dependent normal and shear granular resistive forces. While normal penetration resistance remains nearly unchanged with inclination, shear resistance decreases substantially as slope angle increases. Guided by these measurements, we develop a simple robot-terrain interaction model that predicts anchoring timing, step length, and resulting robot speed, as functions of terrain strength and slope angle. The model reveals that slope-induced performance loss is primarily governed by delayed anchoring and increased backward slip rather than excessive sinkage. By extending the model to generalized terrain conditions, we construct failure phase diagrams that identify sinkage- and slippage-induced failure regimes, enabling quantitative risk estimation for locomotion on granular slopes. This physics-informed framework provides predictive insight into terrain-dependent failure mechanisms and offers guidance for safer and more robust robot operation on deformable inclines. \IEEEoverridecommandlockouts {IEEEkeywords} Biologically-Inspired Robots; Locomotion on Deformable Terrains; Robotics in Hazardous Fields 1 Introduction Sand slopes are widespread in natural environments, including deserts, coastal regions, riverbanks, and planetary surfaces (Fig. 1 a). These terrains are of increasing interest for mobile robots, for applications such as search-and-rescue, environmental monitoring, rural delivery, and extraterrestrial exploration. However, the combination of deformable substrates and steep inclines in these environments pose significant challenges to terrestrial locomotion, where robots often encounter excessive sinkage, slippage, or loss of stability [ 1 , 2 , 3 ] (Fig. 1 b, c). Misapplied locomotion strategies can lead to catastrophic failures and complete immobilization. A mechanistic understanding of how robots interact with sand slopes is therefore essential for assessing traversal risks and extending operational envelopes [ 4 ] (Fig. 1 d). A fundamental challenge of locomotion on sand is that granular substrates can exhibit both solid-like and fluid-like behaviors [ 5 , 6 ] , leading to complex interactions between limb intrusion, force generation, and terrain yielding [ 7 , 8 , 9 , 10 ] . Prior studies of locomotion on level granular media have characterized resistive forces during intrusion and propulsion, leading to models that successfully predict performance on level sand [ 11 , 12 ] . These works established how penetration depth and normal resistive forces regulated step length and speed on level sand. Figure 1: Importance and challenges of legged locomotion on sand slopes. (a) Sand slopes widely present in natural environments (White Sands National Park, NM). (b) Hexapedal robot ascending a dune slope. (c) Quadruped robot traversing a relatively mild sand slope at White Sands. (d) Example ascent paths on a sand slope, illustrating the tradeoff between shorter routes (yellow) and longer but potentially less risky path (red, cyan). On inclined sand, however, gravity alters the internal stress state of the granular medium, reducing available shear resistance and promoting downslope yielding [ 13 , 14 ] . Previous robotic and biological studies have shown that increasing slope angle degrades speed, increases energetic cost, and alters gait stability [ 15 , 16 , 1 ] . Bio-inspired strategies, such as lateral undulation in snakes or specialized limb kinematics in insects, have demonstrated improved performance by mitigating slip and redistributing forces [ 17 , 15 ] . Together, these studies highlight the importance of force generation and terrain yielding in determining locomotor success on granular inclines. Despite this progress, most existing research on robot locomotion on slopes [ 18 , 19 , 1 ] do not directly resolve the underlying force interactions between the robot and the granular medium on slopes. Without linking performance degradation to measurable changes in normal and shear resistance, it remains difficult to determine whether failures arise primarily from excessive sinkage, insufficient shear anchoring, or their interaction. Consequently, predictive understanding of slope-induced failure mechanisms remains limited. To address this gap, in this paper we experimentally measured the locomotion performance of a hexapedal robot on a tiltable sand fluidization bed (Sec. 2 ). By systematically quantifying slope-dependent normal and shear resistances, we identified how granular anisotropy alters limb anchoring dynamics. We then developed a simplified force-balance model linking shear strength to anchoring time, step length, and speed (Sec. 3 ). Finally, we generalized this framework to construct terrain–robot phase diagrams that distinguish sinkage- and slip-dominated failure regimes, enabling model-informed risk prediction for locomotion on granular slopes (Sec. 4 ). 2 Materials and Methods 2.1 Robot To investigate locomotion performance on sand slopes, we developed a simple hexapedal robot (22 cm length \times 10 cm width, 350 grams, Fig. 2 a) to perform controlled locomotion experiments. The sand-climbing hexapod robot were equipped with six 3-D printed C-shaped legs (Fig. 2 a,b). The angular positions of the robot legs was controlled using a microcontroller (Arduino Uno), and actuated by servo motors (Dynamixel XL330). The radius of the C-shaped leg was 4 cm and the width of the leg was 1 cm. During all trials, the robot was programmed to use an alternating tripod gait [ 20 ] in which the legs were divided into 2 groups (front and rear leg from one side, plus the middle leg from the other side), and the two groups alternated with a 180-degree phase difference with each other. This alternating tripod gait is a bio-inspired gait often used by insects [ 21 ] , and was found to exhibit good locomotion performance on leveled sand [ 17 , 11 ] . For all trials, the leg angular speed was kept constant at \omega=2\pi rad/s. Figure 2: Robot and experimental setup for studying locomotion on granular slopes. (a) Hexapedal robot used in this study. (b) Diagram illustrating the key parameters of robot leg. Here R denotes robot leg radius, d denotes the leg penetration depth, \omega denotes the leg angular velocity, and h denotes the hip height, measured as the distance from the leg axle to the bottom of the robot body. (c) Schematic of the tiltable granular trackway for testing robot locomotion on sand. In b and c, \theta denotes the slope angle of the granular incline. 2.2 Robot Locomotion Experiments Robot locomotion experiments were performed in a 127 cm \times 65 cm trackway (Fig. 2 c). The trackway was filled with 300 \mu m diameter glass beads (Grainger) as a model granular substrate, to a depth of 10 cm to minimize boundary effects [ 2 ] . The glass beads have been proven to be a promising model substrate for studying the interaction between robots and granular materials [ 22 ] , while the simpler geometry facilitates the control of the substrate to achieve uniform packing state [ 23 ] . To ensure a consistent initial condition across locomotion tests, the granular bed was fluidized prior to each trial by sending air upward through the base of the trackway. After the air flow slowly ramped down, the sand would settle into a uniform, loosely-packed compaction, eliminating spatial variations in surface height and penetration resistance [ 24 ] . The granular slope angle, \theta , was controlled using two linear actuators, and measured