Abstract
The agricultural sector plays a fundamental role in our society, where it is increasingly important to automate processes, which can generate beneficial impacts in the productivity and quality of products. Perception and computer vision approaches can be fundamental in the implementation of robotics in agriculture. In particular, deep learning can be used for image classification or object detection, endowing machines with the capability to perform operations in the agriculture context. In this work, deep learning was used for the detection of grape bunches in vineyards considering different growth stages: the early stage just after the bloom and the medium stage where the grape bunches present an intermediate development. Two state-of-the-art single-shot multibox models were trained, quantized, and deployed in a low-cost and low-power hardware device, a Tensor Processing Unit. The training input was a novel and publicly available dataset proposed in this work. This dataset contains 1929 images and respective annotations of grape bunches at two different growth stages, captured by different cameras in several illumination conditions. The models were benchmarked and characterized considering the variation of two different parameters: the confidence score and the intersection over union threshold. The results showed that the deployed models could detect grape bunches in images with a medium average precision up to 66.96%. Since this approach uses low resources, a low-cost and low-power hardware device that requires simplified models with 8 bit quantization, the obtained performance was satisfactory. Experiments also demonstrated that the models performed better in identifying grape bunches at the medium growth stage, in comparison with grape bunches present in the vineyard after the bloom, since the second class represents smaller grape bunches, with a color and texture more similar to the surrounding foliage, which complicates their detection.

Abstract
The Trombe wall is a passive solar system that can improve buildings energy efficiency. Despite the studies already developed in this field, more research is needed to assess the possibility of its integration in buildings avoiding user intervention. In this study, the influence of air vent management and materials' heat storage capacity upon its thermal performance, particularly in the temperature fluctuation and indoor conditions, was discussed. Comparing two days with similar solar radiation (SR) for non-ventilated (NVTW) and ventilated (VTW) Trombe walls, a differential of 43 degrees C between the external surface temperature and the one in the middle of the massive wall was verified for NVTW, while for VTW this value was 27 degrees C, reflecting the heat transfer by air convection, which reduced greenhouse effect, solar absorption and heat storage. A cooling capacity greater than 50% was verified for VTW compared to NVTW during night periods. An algorithm for the Trombe wall's automation and control was proposed considering SR as variable. Air vents and external shading devices should be open when SR exceeds 100 W/m(2) and closed for 50 W/m(2) to obtain at least 20 degrees C inside the room. Closing for 50 W/m(2) and opening for values lower that 20 W/m(2) is suggested for summer periods.

Abstract
Robotics navigation and perception for forest management are challenging due to the existence of many obstacles to detect and avoid and the sharp illumination changes. Advanced perception systems are needed because they can enable the development of robotic and machinery solutions to accomplish a smarter, more precise, and sustainable forestry. This article presents a state-of-the-art review about unimodal and multimodal perception in forests, detailing the current developed work about perception using a single type of sensors (unimodal) and by combining data from different kinds of sensors (multimodal). This work also makes a comparison between existing perception datasets in the literature and presents a new multimodal dataset, composed by images and laser scanning data, as a contribution for this research field. Lastly, a critical analysis of the works collected is conducted by identifying strengths and research trends in this domain.

Abstract
Developing ground robots for agriculture is a demanding task. Robots should be capable of performing tasks like spraying, harvesting, or monitoring. However, the absence of structure in the agricultural scenes challenges the implementation of localization and mapping algorithms. Thus, the research and development of localization techniques are essential to boost agricultural robotics. To address this issue, we propose an algorithm called VineSLAM suitable for localization and mapping in agriculture. This approach uses both point- and semiplane-features extracted from 3D LiDAR data to map the environment and localize the robot using a novel Particle Filter that considers both feature modalities. The numeric stability of the algorithm was tested using simulated data. The proposed methodology proved to be suitable to localize a robot using only three orthogonal semiplanes. Moreover, the entire VineSLAM pipeline was compared against a state-of-the-art approach considering three real-world experiments in a woody-crop vineyard. Results show that our approach can localize the robot with precision even in long and symmetric vineyard corridors outperforming the state-of-the-art algorithm in this context.

Abstract
In hydroponic cultivation, monitoring and quantification of nutrients is of paramount importance. Precision agriculture has an urgent need for measuring fertilization and plant nutrient uptake. Reliable, robust and accurate sensors for measuring nitrogen (N), phosphorus (P) and potassium (K) are regarded as critical in this process. It is vital to understand nutrients’ interference; thusly, a Hoagland fertilizer solution-based orthogonal experimental design was deployed. Concentration ranges were varied in a target analyte-independent style, as follows: [N] = [103.17–554.85] ppm; [P] = [15.06–515.35] ppm; [K] = [113.78–516.45] ppm, by dilution from individual stock solutions. Quantitative results for N and K, and qualitative results for P were obtained.

Abstract
There are different clean energy production technologies, including wind energy production. This type of energy, among renewable energies, is one of the least predictable due to the unpredictability of the wind. The wind prediction has been a deeply analysed field since has a considerable share on the green energy production, and the investments on this sector are growing. The efficiency and stability of power production can be increased with a better prediction of the main source of energy, in our case the wind. In this paper, some techniques inspired by Biological Inspired Optimization Techniques applied to wind forecast are compared. The wind forecast is very important to be able to estimate the electric energy production in the wind farms. As you know, the energy balance must be checked in the electrical system at every moment. In this study we are going to analyse different methodologies of wind and power prediction for wind farms to understand the method with best results.