Cutting-edge Large Language Models (LLMs) play a crucial role in improving autonomous navigation by offering efficient solutions. While LLMs require powerful computers to operate, security concerns and maintaining a stable connection with the cloud can be challenging due to various factors. To address this issue, we propose a metareasoning approach for edge-cloud collaborative LLM planning which leads to an efficient autonomous navigation. The proposed approach allows the system to seamlessly switch between cloud and edge devices to fulfill the mission even in the event of a lost connection or entering a GPS-denied environment. Moreover, we deploy state-of-the-art LLM models on resource-constrained systems like the NVIDIA Jetson Orin Nano 8GB, integrated with ROSMASTER X3. These LLMs have demonstrated exceptional utility in dynamic planning for multi-room or maze environments. A comprehensive LLM profiling of TinyLLM models was performed for five different LLMs. The LLM profiling result shows that while certain models with smaller sizes and lower power consumption were available, their accuracy was insufficient for our application requirements. As a result, LLaMa2-7B is considered the edge LLM model due to its optimal balance of performance and accuracy. The experimental results show that under weak signal conditions (<-50 dB), the metareasoning approach improves energy consumption by up to 4x while the cloud-based implementation exceeds the energy consumption of the onboard LLM implementation. Moreover, with delays of 10-20 seconds, cloud implementation becomes impractical for real-time applications in weak signal environments. This underscores the need for metareasoning, which optimizes energy consumption and response time, providing a balanced solution by adapting to signal strength. A real-world implementation of the proposed approach on ROSMASTER X3 with NVIDIA Jetson Orin Nano board can be found in this video which shows that the mission was completed despite losing the connection with cloud-based LLM.

Canonical algorithms for multi-armed bandits typically assume a stationary reward environment where the size of the action space (number of arms) is small. More recently developed methods typically relax only one of these assumptions: existing non-stationary bandit policies are designed for a small number of arms, while Lipschitz, linear, and Gaussian process bandit policies are designed to handle a large (or infinite) number of arms in stationary reward environments under constraints on the reward function. In this manuscript, we propose a novel policy to learn reward environments over a continuous space using Gaussian interpolation. We show that our method efficiently learns continuous Lipschitz reward functions with $\mathcal{O}^*(\sqrt{T})$ cumulative regret. Furthermore, our method naturally extends to non-stationary problems with a simple modification. We finally demonstrate that our method is computationally favorable (100-10000x faster) and experimentally outperforms sliding Gaussian process policies on datasets with non-stationarity and an extremely large number of arms.

The EUMETSAT Satellite Applications Facility for Land Surface Analysis (LSA SAF) generates and disseminates two suites of vegetation products (VEGA) based on SEVIRI/MSG and AVHRR/EPS satellite observations. These suites include Leaf Area Index (LAI), Fractional Vegetation Cover (FVC), and Fraction of Absorbed Photosynthetically Active Radiation (FAPAR). Differences in algorithms and inputs lead to inconsistencies between the biophysical variables obtained from each sensor. This work addresses these inconsistencies by developing a unifying retrieval framework using a hybrid approach that combines radiative transfer models with machine learning. Specifically, a Sparse Multi-output Gaussian Process Regression (SMGPR) is proposed to retrieve multiple variables simultaneously with associated uncertainty estimates. Unlike current MSG algorithms that treat each variable independently, SMGPR jointly derives the three estimates, outperforming both single and multi-output Gaussian Process models in accuracy and scalability. The SMGPR was prototyped using an 18-year data record (2004–2021) from MSG inputs to produce VEGA products at 10-day intervals. A quality assessment, comparing results with LSA SAF estimates and other validated products, showed realistic values and temporal variations across all biomes. The method improved spatio-temporal consistency between MSG and EPS VEGA suites, increasing target accuracy for LAI (73.0% to 75.5%) and FVC (73.4% to 73.8%), and substantially for FAPAR (52.7% to 74.2%). Reduced performance was noted near the edges of the MSG disk, likely due to BRDF input limitations at challenging geometries. This framework is expected to enhance the consistency of vegetation products from current and future EUMETSAT sensors.

The Copernicus Global Land Service (CGLS) ensures operational monitoring of the continental ecosystems through the provision of global terrestrial geophysical variables. The CGLS portfolio covers several satellite-based land products at different spatial and temporal resolution related to energy, vegetation, water, and cryosphere. Before delivering the products to the users a quality assessment is conducted to assure that the products meet the user requirements and to inform the users on the quality, uncertainties and limitations attached to them. The exhaustive quality assessment follows international validation good practices guidelines over an evaluation period (typically 2 years) and is complemented by an operational quality evaluation of the latest delivered products. This paper describes the overall strategy and criteria used to perform the quality assessment of the CGLS land vegetation and water products to reach the highest level of validation and hence confidence in the quality of the CGLS products.

This work describes the different algorithmic steps used to retrieve the first Surface Albedo (SA) product based on Sentinel-3 (S-3) data in the framework of the Copernicus Climate Change Service (C3S). The atmospherically corrected Top-Of-Atmosphere (TOA) reflectances into Top-Of-Canopy (TOC) reflectances are brokered from the Copernicus Global Land Service (CGLS). The TOC reflectances are used to obtain a BRDF model. Next, the spectral and angular integration steps are implemented, which take the latter coefficients as input to produce spectral and broadband albedo quantities. The preliminary quality assessment of BSA broadband albedo for the total shortwave shows good overall spatiotemporal consistency with C3S PROBA-V SA products, and good spatial consistency with MODIS MCD43A3 C6 over the Iberian Peninsula (mean bias lower than 5%), with some underestimation of snow targets.

We present and discuss some recent developments to reach the ultimate validation stage as defined by the Land Product Validation group of the Committee on Earth Observation Systems (CEOS LPV) for vegetation products. We specifically focus on the complementarity between the exhaustive spatial sampling provided by Unmanned Aerial Vehicles (UAVs) or high-resolution satellite data and the exhaustive temporal sampling available with IoTs.

In this work we describe the methodology applied for the retrieval of global LAI, FAPAR and FVC from Advanced Very High Resolution Radiometer (AVHRR) on board the Meteorological-Operational (MetOp) polar orbiting satellites also known as EUMETSAT Polar System (EPS). A novel approach has been developed for the joint retrieval of three parameters (LAI, FVC, and FAPAR) instead of training one model per parameter. The method relies on multi-output Gaussian Processes Regression (GPR) trained over PROSAIL EPS simulations. A sensitivity analysis is performed to assess several sources of uncertainties in retrievals and maximize the positive impact of modeling the noise in training simulations. We describes the main features of the operational processing chain along with the current status of the global EPS vegetation products, including details about its overall quality and preliminary assessment of the products based on intercomparsion with equivalent (MODIS, PROBA-V) satellite vegetation products.

Clumping index (CI) is a measure of foliage aggregation relative to a random distribution of leaves in space. The CI can help with estimating fractions of sunlit and shaded leaves for a given value of leaf area index (LAI). Both the CI and LAI can be obtained from global Earth Observing (EO) systems such as the Moderate Resolution Imaging Spectrometer (MODIS). Here, the compatibility between CI and LAI products derived from EO data is examined independently using the theory of spectral invariants, also referred to as photon recollision probability theory (i.e. ` p-theory'), along with raw LAI-2000/2200 Plant Canopy Analyzer data from 75 sites distributed across a range of plant functional types (PFTs). The p-theory describes the probability (p-value) that a photon, having intercepted an element in the canopy, will recollide with another canopy element rather than escape the canopy. Our results indicate that the integration of empirically-based CI maps with the MODIS LAI product is feasible, providing a potential means to improve the accuracy of LAI EO data products. Given the strong results for the large range of PFTs explored here, we demonstrate the capacity to obtain p-values for any location solely from EO data. This is relevant for future applications of the photon recollision probability concept for global and local monitoring of vegetation using EO data.

Leaf area index (LAI) is a key biophysical parameter used to determine foliage cover and crop growth in environmental studies in order to assess crop yield. LAI estimates can be classified as direct or indirect methods. Direct methods are destructive, time consuming, and difficult to apply over large fields. Indirect methods are non-destructive and cost-effective due to its portability, accuracy and repeatability. In this study, we compare indirect LAI estimates acquired from two classical instruments such as LAI-2000 and digital cameras for hemispherical photography, with LAI estimates acquired with a smart app (PocketLAI) installed on a mobile smartphone. In this work it is shown that LAI estimates obtained with the classical instruments and with a smartphone are well correlated. Consequently, results presented in this work allow considering PocketLAI as a powerful alternative to the classical instruments for LAI monitoring during field campaigns.

In this paper, we propose an approach for estimation biophysical parameters, namely LAI effective, FAPAR, and FCOVER, based on in-situ and satellite measurements. In-situ data were collected during 2013-2014 within several field campaigns at the JECAM test site in Ukraine. We have built 30-meter resolution crop specific maps of biophysical parameters based on regression dependencies between ground measurements and NDVI derived from high resolution imagery (Landsat, SPOT) and Proba-V (100 m). In this paper, we discuss the best model selection for LAI effective, FAPAR and FCOVER mapping as well as selection of optimal source of satellite images. Obtained results are compared to available coarse resolution global biophysical products such as MODIS and SPOT-Vegetation.

Leaf area index (LAI) is identified as a Level 2b product to be derived from the Sentinel-2 (S2) Multispectral Imager (MSI) in support of user services [1]. The Validation of Sentinel 2 (VALSE2) project conducted a review, implementation and validation of LAI algorithms suitable for the MSI. Validation was performed using simulated MSI imagery co-located with in-situ LAI over 7 ESA Campaigns. Here we describe two implemented algorithms, the INRA Neural Network algorithm (NNET) and the CCRS Red-Edge algorithm (CCRS), and report on their verification using the PROSAILH radiative transfer model as well as validation both over the BARRAX ESA Campaign as well as prior campaigns. Results indicate both algorithms can provide reasonably unbiased LAI estimates with acceptable error (1 unit) error over BARRAX. The larger error may be due to a combination of noisy input image data as well as the combination of sparse canopies and bright soils at that site.