Abstract
This paper proposes a camera-to-3D Light Detection And Ranging calibration framework through the optimization of atomic transformations. The system is able to simultaneously calibrate multiple cameras with Light Detection And Ranging sensors, solving the problem of Bundle. In comparison with the state-of-the-art, this work presents several novelties: the ability to simultaneously calibrate multiple cameras and LiDARs; the support for multiple sensor modalities; the calibration through the optimization of atomic transformations, without changing the topology of the input transformation tree; and the integration of the calibration framework within the Robot Operating System (ROS) framework. The software pipeline allows the user to interactively position the sensors for providing an initial estimate, to label and collect data, and visualize the calibration procedure. To test this framework, an agricultural robot with a stereo camera and a 3D Light Detection And Ranging sensor was used. Pairwise calibrations and a single calibration of the three sensors were tested and evaluated. Results show that the proposed approach produces accurate calibrations when compared to the state-of-the-art, and is robust to harsh conditions such as inaccurate initial guesses or small amount of data used in calibration. Experiments have shown that our optimization process can handle an angular error of approximately 20 degrees and a translation error of 0.5 meters, for each sensor. Moreover, the proposed approach is able to achieve state-of-the-art results even when calibrating the entire system simultaneously.

Abstract
Current chemometrics and artificial intelligence methods are unable to deal with complex multi-scale interference of blood constituents in visible shortwave near-infrared spectroscopy point-of-care technologies. The major difficulty is to access the rich information in the spectroscopy signal, unscrambling and interpreting spectral interference to provide analytical quality quantifications. We present a new self-learning artificial intelligence method for spectral processing based on the search of covariance modes with direct correspondence to the BeerLambert law. Dog and cat hemograms were analyzed by impedance flow cytometry and standard laboratory methods (erythrocytes counts, hemoglobin, and hematocrit). Spectral records were performed for the same samples. The methodology was benchmarked against state-of-the-art chemometrics: a multivariate linear model of hemoglobin bands, similarity, partial least squares, local partial least squares, and artificial neural networks. The new method outperforms the state-of-the-art, providing analytical quality quantifications according to desired veterinary pathology guidelines (total errors of 1.69% to 7.14%), whereas chemometric methods cannot. The method finds relevant samples and spectral information that hold the quantitative information for a particular interference mode, in contrast to the current methods that do not hold a relationship with the BeerLambert law. It allows the interpretation of interference bands used in quantification, providing the capacity to determine if the composition of an unknown sample is predictable. This research is especially relevant for improving current optical point-of-care technologies that are affected by spectral interference and moving towards micro-sampling and reagent-less technologies in healthcare and veterinary medicine diagnosis.

Abstract
Software for robotic systems is becoming progressively more complex despite the existence of established software ecosystems like ROS, as the problems we delegate to robots become more and more challenging. Ensuring that the software works as intended is a crucial (but not trivial) task, although proper quality assurance processes are rarely seen in the open-source robotics community. This paper explains how we analyzed and improved a specialized path planner for steep-slope vineyards regarding its software dependability. The analysis revealed previously unknown bugs in the system, with a relatively low property specification effort. We argue that the benefits of similar quality assurance processes far outweigh the costs and should be more widespread in the robotics domain.

Abstract
The importance of agriculture and the production of fruits and vegetables has stood out mainly over the past few years, especially for the benefits for our health. In 2021, in the international year of fruit and vegetables, it is important to encourage innovation and evolution in this area, with the needs surrounding the different processes of the different cultures. This paper compares the performance between two datasets for robotics fruit harvesting using four deep learning object detection models: YOLOv4, SSD ResNet 50, SSD Inception v2, SSD MobileNet v2. This work aims to benchmark the Open Images Dataset v6 (OIDv6) against an acquired dataset inside a tomatoes greenhouse for tomato detection in agricultural environments, using a test dataset with acquired non augmented images. The results highlight the benefit of using self-acquired datasets for the detection of tomatoes because the state-of-the-art datasets, as OIDv6, lack some relevant characteristics of the fruits in the agricultural environment, as the shape and the color. Detections in greenhouses environments differ greatly from the data inside the OIDv6, which has fewer annotations per image and the tomato is generally riped (reddish). Standing out in the use of our tomato dataset, YOLOv4 stood out with a precision of 91%. The tomato dataset was augmented and is publicly available (See https://rdm.inesctec.pt/ and https://rdm.inesctec.pt/dataset/ii-2021-001).

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 world society needs to produce more food with the highest quality standards to feed the world population with the same level of nutrition. Microfarms and local food production enable growing vegetables near the population and reducing the operational logistics costs related to post-harvest food handling. However, it isn't economical viable neither efficient to have one person devoted to these microfarms task. To overcome this issue, we propose an open-source robotic solution capable of performing multitasks in small polyculture farms. This robot is equipped with optical sensors, manipulators and other mechatronic technology to monitor and process both biotic and abiotic agronomic data. This information supports the consequent activation of manipulators that perform several agricultural tasks: crop and weed detection, sowing and watering. The development of the robot meets low-cost requirements so that it can be a putative commercial solution. This solution is designed to be relevant as a test platform to support the assembly of new sensors and further develop new cognitive solutions, to raise awareness on topics related to Precision Agriculture. We are looking for a rational use of resources and several other aspects of an evolved, economically efficient and ecologically sustainable agriculture.