Estimating above-ground biomass (AGB) is essential for monitoring vegetation and assessing carbon stock. Traditional field-based methods, although accurate, are costly and difficult to implement in densely vegetated environments. In this context, the use of UAVs equipped with LiDAR sensors has emerged as a promising solution for forest mapping. This study integrates Coverage Path Planning (CPP) algorithms with an autonomous navigation system that uses obstacle avoidance, enabling efficient exploration of unknown areas. Three CPP approaches were evaluated: Polygon Coverage, Grid Coverage, and Spiral Coverage. The experimental results indicate that combining the Grid Coverage method with autonomous navigation enhances forest area coverage, reduces execution time, and optimizes data acquisition for biomass estimation.
Interaction control has become increasingly important with the rise of human-robot collaboration. Among various strategies, impedance and admittance control stand out by regulating the relationship between velocity and force, rather than individual system states. Impedance control offers stability in stiff environments but may lack precision in free space, while admittance control performs well in non-contact tasks but can be unstable with stiff environments. Since stiff environments exhibit high admittance, impedance control is generally preferred, and vice versa. To overcome the limitations of each method, hybrid systems combining both approaches have been developed, typically categorized as switch-based or non-switch-based. This study proposes a comparative framework for such hybrids using a novel visualization method: the impedance space. New metrics are introduced to assess how well these systems achieve the target impedance, enabling quantitative performance evaluation across varying environmental conditions.
This paper presents the case study of admittance controller with dynamic compensator and impedance modulation for a Robotic Cane. The admittance controller provides easy control of the device without user instrumentation. Dynamic compensation corrects for the delay of the device’s dynamics. Impedance modulation provides the device with a fast and soft response. A female post-surgical user, 1.57m in height, 56kg in weight, and 32 years of age, participates in the experiments, which were carried out on a straight road. The support by Robotic Cane was evaluated applying the NASA TLX, the SUS and one adapted questionnaire. The responses obtained from the experimental test show that the control strategy presented in this paper is appropriate for providing the support required during the rehabilitation stage following knee surgery. The results show a low work load requirement (26.25%), high frequency of use (82.5%), and adequate perceived quality of the user experience (73.3%).
Learning from Demonstration (LfD) is a promising approach to teaching robotic manipulators by example, enabling intuitive and efficient behavior acquisition. However, most existing methods struggle with data efficiency and generalization, particularly in scenarios where only a few demonstrations are available. In this work, we propose an architecture that combines Task-Parameterized Gaussian Mixture Models (TP-GMM) with Bayesian Optimization (BO) for automatic hyperparameter tuning. TP-GMM enables the encoding of demonstrations in multiple local reference frames, capturing the relative dynamics of the task and facilitating generalization to unseen configurations. BO, in turn, automatically selects key model parameters, such as the number of components, covariance regularization, and input-output weights, by maximizing model performance via cross-validation. We validate the proposed approach in a real-world pick-and-place manipulation task using a Kinova Gen3 robot, with varying task parameters inferred through visual markers. By relying on only five demonstrations per subtask, our method achieved a 95% task success rate, outperforming both standard TP-GMM and a TP-GMM+EKF variant in terms of accuracy and consistency. These results demonstrate that our method outperforms non-parameterized models in accuracy and robustness while maintaining low computational cost and high sample efficiency, highlighting its potential for scalable and generalizable robotic learning.
This paper presents an enhanced version of the Blockly–ROS platform for mobile robot control, expanding its capabilities with loop structures and a gesture recognition module for Brazilian Sign Language (Libras). The proposed system integrates Blockly’s block-based visual programming with ROS 2 middleware, enabling both iterative logic and conditional control through real-time gesture detection using a YOLOv8n-based model. The platform was validated through a playful tic-tac-toe game scenario, where the robot autonomously navigated the board and executed movements based on hand-signed letters from the user. Experimental results demonstrate high accuracy in gesture recognition and precise trajectory execution, highlighting the potential of this approach for inclusive and interactive education. In particular, the system fosters playful Libras learning, motor laterality training, and the integration of computer vision with robotic control, offering a low-cost, accessible, and pedagogically robust environment for diverse educational contexts.
