Stefano Carpin’s research while affiliated with University of Milan and other places

The advent of machine learning technologies in conjunction with the advancements in UAV-based remote sensing pioneered a new era of research in agriculture. The escalating concern for water management in drought-prone areas such as California underscores the urgent need for sustainable solutions. Stem water potential (SWP) measurement using pressure chambers is one of the most common methods used to directly determine tree water status and the optimal timing for irrigation in orchards. However, this approach is inefficient due to its labor-intensive nature. To address this problem, we used weather, thermal and multispectral data as inputs to the machine learning (ML) algorithms to predict the SWP of pistachio and almond trees. For each crop, we first deployed six supervised ML classification models: Random Forest (RF), Support Vector Machine (SVM), Gaussian Naive Bayes (GNB), Decision Tree (DT), K-Nearest Neighbors (KNN), and Artificial Neural Network (ANN). All classifiers provided more than 79% of accuracy while RF showed high performance in both pistachio and almond orchards at 88% and 89%, respectively. The feature importance results by the RF model revealed that the weather features were the most influential factors in the decision-making process. In both crops, canopy temperature 𝑇𝑐 was the next important feature closely followed by OSAVI in pistachios and NDVI in almonds. RF regression model predicted SWPs with 𝑅2 of 0.70 in pistachio and 𝑅2 of 0.55 in the almond orchard. Our results demonstrate that ML models are practical tools for irrigation scheduling decisions. This study offered a data-driven approach that effectively balances minimal data requirements with accuracy to facilitate optimal water management for end-users.

Leveraging the power of a graph neural network (GNN) with message passing, we present a Monte Carlo Tree Search (MCTS) method to solve stochastic orienteering problems with chance constraints. While adhering to an assigned travel budget the algorithm seeks to maximize collected reward while incurring stochastic travel costs. In this context, the acceptable probability of exceeding the assigned budget is expressed as a chance constraint. Our MCTS solution is an online and anytime algorithm alternating planning and execution that determines the next vertex to visit by continuously monitoring the remaining travel budget. The novelty of our work is that the rollout phase in the MCTS framework is implemented using a message passing GNN, predicting both the utility and failure probability of each available action. This allows to enormously expedite the search process. Our experimental evaluation shows that with the proposed method and architecture we manage to efficiently solve complex problem instances while incurring in moderate losses in terms of collected reward. Moreover, we demonstrate how the approach is capable of generalizing beyond the characteristics of the training dataset. The paper's website, open-source code, and supplementary documentation can be found at ucmercedrobotics.github.io/gnn-sop.

We present a new Monte Carlo Tree Search (MCTS) algorithm to solve the stochastic orienteering problem with chance constraints, i.e., a version of the problem where travel costs are random, and one is assigned a bound on the tolerable probability of exceeding the budget. The algorithm we present is online and anytime, i.e., it alternates planning and execution, and the quality of the solution it produces increases as the allowed computational time increases. Differently from most former MCTS algorithms, for each action available in a state the algorithm maintains estimates of both its value and the probability that its execution will eventually result in a violation of the chance constraint. Then, at action selection time, our proposed solution prunes away trajectories that are estimated to violate the failure probability. Extensive simulation results show that this approach can quickly produce high-quality solutions and is competitive with the optimal but time-consuming solution.

The ROS 2 Navigation Stack (Nav2) has emerged as a widely used software component providing the underlying basis to develop a variety of high-level functionalities. However, when used in outdoor environments such as orchards and vineyards, its functionality is notably limited by the presence of obstacles and/or situations not commonly found in indoor settings. One such example is given by tall grass and weeds that can be safely traversed by a robot, but that can be perceived as obstacles by LiDAR sensors, and then force the robot to take longer paths to avoid them, or abort navigation altogether. To overcome these limitations, domain specific extensions must be developed and integrated into the software pipeline. This paper presents a new, lightweight approach to address this challenge and improve outdoor robot navigation. Leveraging the multi-scale nature of the costmaps supporting Nav2, we developed a system that using a depth camera performs pixel level classification on the images, and in real time injects corrections into the local cost map, thus enabling the robot to traverse areas that would otherwise be avoided by the Nav2. Our approach has been implemented and validated on a Clearpath Husky and we demonstrate that with this extension the robot is able to perform navigation tasks that would be otherwise not practical with the standard components.

In multi-robot informative path planning the problem is to find a route for each robot in a team to visit a set of locations that can provide the most useful data to reconstruct an unknown scalar field. In the budgeted version, each robot is subject to a travel budget limiting the distance it can travel. Our interest in this problem is motivated by applications in precision agriculture, where robots are used to collect measurements to estimate domain-relevant scalar parameters such as soil moisture or nitrates concentrations. In this paper, we propose an online, distributed multi-robot sampling algorithm based on Monte Carlo Tree Search (MCTS) where each robot iteratively selects the next sampling location through communication with other robots and considering its remaining budget. We evaluate our proposed method for varying team sizes and in different environments, and we compare our solution with four different baseline methods. Our experiments show that our solution outperforms the baselines when the budget is tight by collecting measurements leading to smaller reconstruction errors.

Motivated by our ongoing work in robotics for precision agriculture, in this work we consider the problem of estimating a scalar field using a team of robots collecting samples and subject to a travel budget. Our fully distributed method leverages the underlying properties of Gaussian Process regression to promote dispersion using minimal information sharing. Extensive simulations demonstrate that our proposed solution outperforms alternative approaches.