cs.ROFeb 27, 2026

Evaluating Accuracy of Vine Robot Shape Sensing with Distributed Inertial Measurement Units

Authors: Alexis E. Laudenslager, Antonio Alvarez Valdivia, Nathaniel Hanson, Margaret McGuinness

Organizations: Department of Aerospace and Mechanical Engineering, University of Notre Dame, Notre Dame, Indiana, USA. · Lincoln Laboratory, Massachusetts Institute of Technology, Lexington, Massachusetts, USA

Read PDF
Ask AI
Cite

Copy citation

DOINot available for this paper

Abstract

Soft, tip-extending vine robots are well suited for navigating tight, debris-filled environments, making them ideal for urban search and rescue. Sensing the full shape of a vine robot's body is helpful both for localizing information from other sensors placed along the robot body and for determining the robot's configuration within the space being explored. Prior approaches have localized vine robot tips using a single inertial measurement unit (IMU) combined with force sensing or length estimation, while one method demonstrated full-body shape sensing using distributed IMUs on a passively steered robot in controlled maze environments. However, the accuracy of distributed IMU-based shape sensing under active steering, varying robot lengths, and different sensor spacings has not been systematically quantified. In this work, we experimentally evaluate the accuracy of vine robot shape sensing using distributed IMUs along the robot body. We quantify IMU drift, measuring an average orientation drift rate of 1.33 degrees/min across 15 sensors. For passive steering, mean tip position error was 11% of robot length. For active steering, mean tip position error increased to 16%. During growth experiments across lengths from 30-175 cm, mean tip error was 8%, with a positive trend with increasing length. We also analyze the influence of sensor spacing and observe that intermediate spacings can minimize error for single-curvature shapes. These results demonstrate the feasibility of distributed IMU-based shape sensing for vine robots while highlighting key limitations and opportunities for improved modeling and algorithmic integration for field deployment.

Figures & tables

Fig. 1: Summary of contributions. We quantified the shape sensing accuracy of our vine robot instrumented with inertial measurement unit (IMU) sensors along its length at various (a) passive and (b) active steering angles, (c) robot lengths, and (d) sensor spacings. We also measured the effect on individual IMU orientation accuracy of drift over time.
Fig. 2: Vine robot shape sensing system. The vine robot includes a main chamber, three series pouch motor (SPM) steering actuators, and a base for growth and retraction. Eighteen IMUs are distributed along the robot body and connected through multiplexers to a microcontroller at the base.
Fig. 3: Orientation offsetting procedure. Yellow reference frames represent the orientation of each IMU i , and red reference frames represent the orientation of IMU i-1 , the previous IMU (closer to the base). (a) Raw IMU orientations when the robot is straight. (b) After offsetting, consecutive IMU orientations are aligned by subtracting the stored relative orientation measured in the straight configuration.
Fig. 4: Vine robot model parameters used for shape sensing algorithm. Consecutive IMUs, indexed ii and i+1i+1 , define a local segment of length ss with inflated diameter dd . An orientation difference θ\theta between IMUs is modeled as a single bend located midway between them, consisting of a circular arc of length LarcL_{arc} and two straight segments of length LstraightL_{straight} .
Fig. 5: IMU drift test results. Orientation error relative to the initial offset shown as a function of time since offsetting. Each color corresponds to one of 15 IMUs, with one representative trial per unit. The black line indicates a linear fit across all data ( 1.33∘1.33^{\circ} /min, R2R^{2} = 0.16, pp =0.0024).
Fig. 6: Passive steering test setup and results. (a) At full length (1.73 m), the vine robot was bent against a wooden plank to prescribed angles between 0∘0^{\circ} and 90∘90^{\circ} in 15∘15^{\circ} increments. Two trials were conducted at each angle. (b) Tip position error as a percentage of the robot length, computed using an overhead photo as ground truth, plotted versus bend angle. No statistically significant relationship between bend angle and tip position error was observed ( p=0.8p=0.8 ). On average, the error is approximately 11% of the robot length.
Fig. 7: Active steering test setup and results. (a) At full length (1.73 m), a single series pouch motor was inflated to various pressures to induce various curvatures while the robot remained on a flat surface. (b) Tip position error as a percentage of the robot’s length plotted as a function of curvature. Each point represents one trial. No statistically significant relationship between curvature and tip position error was observed ( p=0.38p=0.38 ). On average, the error is approximately 16% of the robot length.
Fig. 8: Length test setup and results. (a) Vine robot grown to variable lengths while actively steered. (b) Tip position error (percentage of robot length) plotted as a function of robot length. Each point represents one trial. A positive trend between length and error was observed, but the relationship did not reach statistical significance ( pp = 0.10). The average error across trials was approximately 8%8\% of robot length.
Fig. 9: Sensor spacing test setup and results. IMU data from two passive and two active steering trials were reanalyzed while considering only sensors spaced a distance ss apart. (a) Estimated robot centerlines for five sensor spacings ranging from 10.2 to 162.6 cm. (b) Tip position error (percentage of robot length) as a function of sensor spacing. While one passive trial achieved its lowest error at the smallest spacing, the remaining three trials exhibited minimum error at intermediate spacings (40–80 cm), indicating that denser sensing did not consistently yield lower error.
ExperimentMean Tip ErrorSignificance
Passive steering11% of Lengthp=0.8p=0.8
Active steering16% of Lengthp=0.38p=0.38
Length test8% of Lengthp=0.10p=0.10
Drift rate1.33 ∘ /minp=0.0024p=0.0024
Sensor spacingIntermediate best—
TABLE I: Summary of Experimental Findings
More in this topic

Continuum Robot Shape Estimation

See all 17 papers

Explore similar work

CardsList
Jun 4, 2026

Gotta Grow Fast: Design and Benchmarking of a Tip Mount for High-Speed Vine Robots

Soft, growing vine robots extend through tip eversion, a mechanism that enables navigation through cluttered environments. However, integrating cameras and other sensors at the tip is uniquely challenging because the material forming the tip is constantly renewed as the robot grows. This continual material turnover, combined with friction between internal layers, added tip weight, and fabric constriction, complicates sensor and tool mounting. These limitations hinder the deployment of vine robots for inspection and search tasks, where rapid growth while carrying tip-mounted sensors is essential. In this work, we present a triangular roller tip mount that reduces internal resistance during growth by rolling rather than sliding against the robot body. The design was refined through iterative failure analysis, enabling, for the first time, consistent eversion on a TPU-coated ripstop nylon vine robot. To quantitatively evaluate mount performance, we introduce a custom testbed that isolates tip mounting effects by measuring tail tension during eversion. Comparative experiments across multiple mount variants, including prior designs, show that our triangular roller mount achieves the lowest tail tension and most repeatable growth performance. These results establish both a validated tip mount design and a repeatable benchmarking framework for advancing sensor and tool integration in soft growing robots. CAD for the mount and testbed is available at: https://sprout-mitll.github.io/tip_mounts/.

Antonio Alvarez Valdivia, Robert Reeve, Ankush Dhawan +4

Lincoln Laboratory, Massachusetts Institute of Technology, Lexington, Massachusetts, USA

RoboticsSoft Robotics
AskPDF
Jun 22, 2026

PanoVine: Whole-Body Visuomotor Control for Soft Growing Vine Robot

Vine robots, a class of soft, growing robots, are suitable for navigating complex and confined environments due to their compliant bodies and self-supporting growth mechanism. However, hysteresis, tether interactions, and deformations make them difficult to predict and model, which in turn limits the effectiveness of conventional planning and control approaches. In this work, we present a data-driven, vision-based control framework for the first autonomous vine robot system. Our system integrates 19 cameras distributed along the robot's body to provide comprehensive feedback of both the robot state and the surrounding environment. Using this rich whole-body vision feedback, we train an end-to-end visuomotor policy from demonstrations for closed-loop autonomous control in complex environments. The policy efficiently aggregates information from distributed sensing while maintaining robustness to inaccurate robot states and actuation. Experimental results demonstrate that the learned policy enables robust navigation and manipulation in challenging scenarios, including steering through branched structures, climbing up slopes, traversing unsupported terrain, reaching objects precisely, and maneuvering through confined spaces and obstacles. Project website https://panovine-bot.github.io

Yimeng Qin, Xiaomeng Xu, William Heap +3

Stanford University

Robot NavigationVisuomotor Policy LearningRobotic ManipulationContinuum Robot ControlSoft Robotics
AskPDF
Jul 22, 2026

Distributed Acoustic Localization Array Deployed Using a Soft Everting Vine Robot

Soft robot exteroception is increasingly being explored for a variety of field applications. In this work, we present a sound-based system for localizing disaster victims in confined and unstructured environments, based on a distributed acoustic sensing architecture embedded along the body of a soft everting vine robot. We propose a dynamic Steered Response Power with Phase Transform framework that supports both far-field direction-of-arrival estimation and near-field three-dimensional source localization as the robot approaches the sound source. To better understand the design and control space related to localizing sound using a soft, shape-morphing robot body, we conduct experiments measuring the accuracy of these methods for a five-microphone array attached to the robot body using three placements relative to the outer membrane of the robot (inside the pressurized body, inside the inner tail, and outside the outer wall) and in four robot configurations (linear, double linear, circular, and sinusoidal). We measure the change in accuracy as the signal-to-noise ratio, the direction of approach, and the distance of the sound source from the center of the array change. Finally, we demonstrate a vine robot growing into an arbitrary shape while carrying microphones along its outer wall, and show that a sound source located with the array's near field can be localized with high accuracy after only three microphones have everted from the robot body. These results highlight the potential of distributed acoustic sensing for reliable victim localization using soft growing robots.

Sebastian Lorca Godoy, Ciera McFarland, Michael Val +3

Department of Aerospace and Mechanical Engineering, University of Notre Dame, Notre Dame, IN 46556, USA. · Pontifical Catholic University of Chile, 8320165 Santiago, Región Metropolitana, Chile. · Lincoln Laboratory, Massachusetts Institute of Technology, Lexington, MA 02421, USA.

Direction-of-Arrival EstimationRoboticsDistributed Acoustic SensingSearch-and-Rescue RoboticsSoft Robotics
AskPDF