Autonomous robots operating in natural environments must often interact with vegetation rather than simply avoid it. In this context, traversability is typically defined from the robot's perspective, by measuring how a specific platform responds when moving through the environment. While practical, this viewpoint entangles the assessment of the environment with the robot's own dynamics, making the resulting characterization difficult to transfer across different platforms. More importantly, it does not directly reflect the properties of the vegetation itself, which are the true source of interaction and potential damage in applications such as agriculture and environmental monitoring. To address this limitation, we propose to characterize vegetation through its intrinsic mechanical properties, independently of any specific robot. By combining deformation measurements with contact force data, we estimate the underlying mechanical parameters and reconstruct the vegetation's response to interaction. This enables vegetation-aware navigation based on intrinsic environmental properties rather than platform-dependent metrics.
Figures & tables
Figure 1: A mobile robot facing a patch of mixed vegetation, deciding whether to push through it or go around. From appearance alone, the ivy, brambles, and woody stems are difficult to distinguish, yet they differ in how much they resist being pushed through.
Figure 2: Overview of the rod-theoretic modeling pipeline. The RGB-D stream is segmented and reduced to a centerline and curvature κ(s) , while the contact loading Fc and initial contact height h are supplied to the Elastic Rod Model, which returns the internal moment M(s) , and the base torque τ . The spatial-profile branch (top) combines κ(s) and M(s) into the flexural-rigidity profile EI(s) ; the lumped-stiffness branch (bottom) recovers the deflection angle θ(t) from the contact location (xc(t),zc(t)) and combines it with τ to give the rotational stiffness kθ . EI(s) and kθ are two independent outputs of the same underlying model, and the choice between them depends on the intended downstream application.
Figure 3: Both panels show a tapered stem of length L , clamped at its base, under a horizontal contact force Fc at height h , point P . (a) The Kirchhoff rod bends continuously, with contact point P=r(sc) and tangent angle θ(s) varying along the rod. (b) The spring rod reduces the stem to a rigid link on an equivalent spring kθ , representing its distributed compliance below P : the same force gives one deflection angle θ and base torque τ .
Figure 4: Experimental specimens pushed by the wire sensor of [ 16 ] , with the segmentation mask (blue) and extracted centerline (dashed yellow) overlaid. (a) Artificial grass bush. (b) Woody twig collected from the wild.
Figure 5: Flexural-rigidity EI(s) against arc length s for the twig (left) and the grass (right). The black dashed curve is the pooled direct estimate. The solid curve is the fitted taper law ( 9 ): red for the twig, magenta for the grass. The remaining curves show EI(s) for taper exponents p spanning a range around the fitted value.
Figure 6: Contact force (median ± IQR per angle bin) as a function of the deflection angle θ (x-axis), with the torque-balance fit (dashed), at three contact heights per specimen. The fitted slope corresponds to the rotational stiffness kθ of the lumped-stiffness model.
Figure 7: Height-variance of the lumped rotational stiffness kθ from the spring-rod model (top) against the distributed flexural rigidity EI(s) from the Kirchhoff-rod model (bottom), both evaluated at the contact height h of each push.
This paper presents TRACE, a traversability-aware reactive navigation framework designed for the autonomous operation of ground robots in unstructured forest environments. Unlike traditional approaches, TRACE leverages VLM-based confidence maps from positive and negative natural-language prompt sets and combines them through evidential fusion to produce a dense traversability map that prioritizes the terrain based on signs of previously traversed paths. Traversability-annotated points are then extracted and used to derive the left and right boundaries of the traversable path via percentile binning followed by spline or regularized polynomial fitting. These boundaries are then incorporated as soft barrier constraints within an NMPC formulation, enabling safe and adaptive path following in complex natural environments. Rigorous experimental trials carried out in real-life forest environments demonstrate TRACE ability for smooth navigation in cluttered, deformable, and visually ambiguous terrains.
Mario A. V. Saucedo, Midhun Edatharam Kunnath, Elias Small +4
Robotics and AI Team, Lule˚a University of Technology, Lule˚a, SE-97187, Sweden. · Korea Advanced Institute of Science & Technology (KAIST), Daejeon, 34141, South Korea.
Terrain assessment is a critical capability for off-road mobile robots, enabling safe and reliable navigation through unstructured and geometrically complex environments. Conventional geometry-based terrain assessment is fast to compute but often overly conservative in unstructured environments. We present PIVOT: a Physically Informed Vision-Language Off-Road Traversability navigation system that augments conventional geometry-based planning with vision-language-model (VLM)-based semantic reasoning for field robots. To physically ground this assessment, we quantify how strongly the VLM's predicted traversal energy cost, robot vibration, and wheel slip correlate with real-world measurements and introduce a unified traversability score that weights each modality by its prediction-measurement correlation. For efficiency, we design a two-level navigation architecture that retains geometry-based planning as the nominal mode and invokes semantic replanning only when that mode fails to find a path. Across five repeated closed-loop trials on a mixed-terrain route totalling around 6.4 km, the proposed system increases overall autonomy from 59.6% to 97.0%, reduces human interventions from 11 to 3, and increases the mean distance between interventions (MDBI) from 69.2 m to 412.9 m compared with geometry-only navigation. These results demonstrate that physically grounded VLM-based terrain assessment can substantially extend autonomous navigation beyond the limitations of geometry alone, while preserving efficient geometric planning as the nominal mode.
Agriculture faces many challenges, and robotic systems can play an important role in addressing them by improving the efficiency and sustainability of field operations. Among these challenges, preserving soil health is a critical concern, as vehicle-soil interactions can degrade the soil structure and produce unwanted surface deformation. A key step toward soil-aware robotics is to explicitly account for how vehicle traffic deforms the ground, yet soil state is typically not treated as a variable. We address this gap by proposing a framework to quantify traffic-induced soil deformation and estimate its evolution online from lidar observations. The method relies on a reduced-order parametric model that represents the soil behavior via physically interpretable parameters, yielding a continuously updated and observable representation of soil state. Experiments conducted in different soil conditions demonstrate the ability of the approach to capture deformation induced by the robot. By making soil response measurable and interpretable during operation, the proposed framework establishes a basis for soil-aware robotic operation, in which the estimated state can be exploited to adapt robotic behaviors in order to reduce soil degradation.
Tom Montagnon, Johann Laconte, Benoit Thuilot +2
Universit´e Clermont Auvergne, INRAE, UR TSCF, 63000, Clermont-Ferrand, France · Institut Pascal, Universit´e Clermont Auvergne, Clermont Auvergne INP, CNRS, 63000 Clermont-Ferrand, France · National Agriculture and Food Research Organization, 1-31-1 Kannondai, Tsukuba, Ibaraki 305-0856, Japan