Sandro Augusto Magalhães

Abstract

Abstract
The development of robotic solutions for agriculture requires advanced perception capabilities that can work reliably in any crop stage. For example, to automatise the tomato harvesting process in greenhouses, the visual perception system needs to detect the tomato in any life cycle stage (flower to the ripe tomato). The state-of-the-art for visual tomato detection focuses mainly on ripe tomato, which has a distinctive colour from the background. This paper contributes with an annotated visual dataset of green and reddish tomatoes. This kind of dataset is uncommon and not available for research purposes. This will enable further developments in edge artificial intelligence for in situ and in real-time visual tomato detection required for the development of harvesting robots. Considering this dataset, five deep learning models were selected, trained and benchmarked to detect green and reddish tomatoes grown in greenhouses. Considering our robotic platform specifications, only the Single-Shot MultiBox Detector (SSD) and YOLO architectures were considered. The results proved that the system can detect green and reddish tomatoes, even those occluded by leaves. SSD MobileNet v2 had the best performance when compared against SSD Inception v2, SSD ResNet 50, SSD ResNet 101 and YOLOv4 Tiny, reaching an F1-score of 66.15, an mAP of 51.46 and an inference time of 16.44ms with the NVIDIA Turing Architecture platform, an NVIDIA Tesla T4, with 12 GB. YOLOv4 Tiny also had impressive results, mainly concerning inferring times of about 5ms.

Abstract

Abstract
A robots simulation system is a basis need for any robotics application. With it, developers teams of robots can test their algorithms and make initial calibrations without risk of damage to the real robots, assuring safety. However, build these simulation environments is usually a time-consuming work, and when considering robot fleets, the simulation reveals to be computing expensive. With it, developers building teams of robots can test their algorithms and make initial calibrations without risk of damage to the real robots, assuring safety. An omnidirectional robot from the 5DPO robotics soccer team served to test this approach. The modeling issue was divided into two steps: modeling the motor's non-linear features and modeling the general behavior of the robot. A proper fitting of the robot was reached, considering the velocity robot's response. © 2020 IEEE.

Abstract
Robotic platforms are being developed for precision agriculture, to execute repetitive and long term tasks. Autonomous monitoring, pruning, spraying and harvesting are some of these agricultural tasks, which requires an advanced path planning system aware of maximum robot capabilities (mobile platform and arms), terrain slopes and plant/fruits position. The state of the art path planning systems have two limitations: are not optimized for large regions and the path planning is not aware of agricultural tasks requirements. This work presents two solutions to overcome these limitations. It considers the VGR2TO (Vineyard Grid Map to Topological) approach to extract from a 2D grid map a topological map, to reduce the total amount of memory needed by the path planning algorithm and to reduce path search space. Besides, introduces an extension to the chosen algorithm, the Astar algorithm, to ensure a safe path and a maximum distance from the vine trees to enable robotic operations on the tree and its fruits. © 2019 IEEE.