Searcharxiv⌕ Search
Home/Archive/arXiv

arXiv · 2309.05128

Robot-assisted Soil Apparent Electrical Conductivity Measurements in Orchards

Dimitrios Chatziparaschis·Elia Scudiero·Konstantinos Karydis
View original source ↗Read source document ↗Export metadata ↓

Abstract

Soil apparent electrical conductivity (ECa) is a vital metric in Precision Agriculture and Smart Farming, as it is used for optimal water content management, geological mapping, and yield prediction. Several existing methods seeking to estimate soil electrical conductivity are available, including physical soil sampling, ground sensor installation and monitoring, and the use of sensors that can obtain proximal ECa estimates. However, such methods can be either very laborious and/or too costly for practical use over larger field canopies. Robot-assisted ECa measurements, in contrast, may offer a scalable and cost-effective solution. In this work, we present one such solution that involves a ground mobile robot equipped with a customized and adjustable platform to hold an Electromagnetic Induction (EMI) sensor to perform semi-autonomous and on-demand ECa measurements under various field conditions. The platform is designed to be easily re-configurable in terms of sensor placement; results from testing for traversability and robot-to-sensor interference across multiple case studies help establish appropriate tradeoffs for sensor placement. Further, a developed simulation software package enables rapid and accessible estimation of terrain traversability in relation to desired EMI sensor placement. Extensive experimental evaluation across different fields demonstrates that the obtained robot-assisted ECa measurements are of high linearity compared with the ground truth (data collected manually by a handheld EMI sensor) by scoring more than $90\%$ in Pearson correlation coefficient in both plot measurements and estimated ECa maps generated by kriging interpolation. The proposed robotic solution supports autonomous behavior development in the field since it utilizes the ROS navigation stack along with the RTK GNSS positioning data and features various ranging sensors.

Explore related subjects

cs.RO

Keep this discovery

BibTeXRIS

Dimitrios Chatziparaschis, Elia Scudiero, Konstantinos Karydis. 2023-09-10. Robot-assisted Soil Apparent Electrical Conductivity Measurements in Orchards. https://arxiv.org/abs/2309.05128

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

IMLE-VLA: Fast Single-Step Action Generation for Vision-Language-Action Policies

Vision-language-action (VLA) policies leverage pretrained vision-language backbones to achieve strong cross-task generalization. A leading design couples this backbone with a dedicated continuous action head trained via diffusion or flow matching. However, such heads rely on iterative multi-step sampling, for example 10 Euler steps in $\pi_{0.5}$. This creates an inference bottleneck that produces stop-and-go movement in the robot and slower task completion. We introduce IMLE-VLA, which replaces the iterative action head with a single-step conditional generator trained via conditional Implicit Maximum Likelihood Estimation (cIMLE). The cIMLE objective promotes multimodal action coverage, avoiding the mode collapse of naive regression heads while eliminating multi-step sampling entirely. When IMLE-VLA is applied to $\pi_{0.5}$, it increases inference frequency 3.67x (55 Hz vs. 15 Hz), enabling up to 11x higher action throughput. On the 40-task LIBERO benchmark, IMLE-VLA achieves the highest average success rate (98.0%) among all baselines while leading in inference frequency. Under the test-time perturbations of LIBERO-plus, IMLE-VLA retains $\pi_{0.5}$'s robustness while other baselines degrade sharply, confirming that the cIMLE head preserves generalization. Real-world experiments on a Franka Emika Panda across four tasks demonstrate smoother motion (2.2x to 3.0x lower jerk) and faster task completion, with IMLE-VLA outperforming $\pi_{0.5}$ on every task and reducing average VLA inference time per episode by 3.9x to 6.6x. Videos and code are available at https://kianhk6.github.io/IMLE-VLA/

cs.RO↗
arXiv2026

ObstaDiff: Generalizable Diffusion Policy Learning via Obstacle-aware Representations

Imitation learning has achieved impressive results in robotic manipulation, yet most existing approaches assume clean backgrounds and lack explicit mechanisms for obstacle-aware motion generation. Extending such policies to cluttered, real-world scenes with unstructured obstacles remains a key generalization challenge. We present ObstaDiff, a decomposed diffusion-policy framework with a lightweight obstacle-aware visual encoder. ObstaDiff extracts a structured target-obstacle-background representation, enabling the downstream alignment policy to generate end-effector trajectories toward a target-centered bottleneck pose while reasoning about surrounding obstacles. We evaluate ObstaDiff on 61 real-robot greenhouse trials per method (366 executions in total). ObstaDiff achieves 75.41% average task success and 8.20% average obstacle collision rate, outperforming representative imitation-learning baselines and improving generalization in cluttered agricultural scenes.

cs.RO↗
arXiv2026

Testing Between the Test Cases: Proving End-to-End Steering in Conditions You Never Drove

AI-based automated vehicle testing is challenging because a model that passes every test condition can still fail in the real world. Formal verification offers a way to directly address this gap. On a simulated highway and an arterial road we trained two small end-to-end steering networks each in CARLA, one on clear conditions alone and one on clear, fog, night and low sun. All four models were driven against a 2.19 ft lane-departure budget. Without driving again, we used bound propagation, a formal method that reads the trained weights, to compute how far steering can drift at every disturbance strength between two captured images. One calculation covers more than a campaign could drive: on the arterial it spans 133 poses, where ten intensities each would be 10^133 combinations, in minutes on one GPU. Not only did formal verification find conditions that broke the clear-trained policy without simulation testing, it provided some preliminary evidence for potential failures between the test cases. Our overall conclusion is that formal verification is a viable complement to simulation, and could be adopted as a part of verification and validation for automated driving.

cs.RO↗