Loading...
3 results
Search Results
Now showing 1 - 3 of 3
- Computational simulation of an agricultural robotic rover for weed control and fallen fruit collection : algorithms for image detection and recognition and systems control, regulation and commandPublication . Ribeiro, João P.L.; Gaspar, Pedro Dinis; Soares, V.N.G.J.; Caldeira, J.M.L.P.The continuous rise in the world’s population has increased the need for food, resulting in a rise of agricultural holdings to ensure the supply of these goods directly to the populations and indirectly to all processing industries in the food business. This situation has led agriculture to reinvent itself and introduce new technics and tools to ensure tighter control of the crops and increase yields in food production. However, the lack of labor coupled with the evolution of weeds resistant to herbicides created a crisis in agricultural food production. However, with the growing evolution in electronics, automation, and robotics, new paths are emerging to solve these problems. A robotic rover was designed to optimize the tasks of weed control and collection of fallen fruits of an orchard. In weed control, a localized spraying system is proposed, therefore reducing the amount of applied herbicides. With fruit collection, it is possible to direct fallen fruits for animal feeding and possible to reduce microbial activity on the next campaign crops, therefore avoiding damage. This study proposes the simulation of this robotic rover on robotic simulation software. It also proposes the replication of a similar environment of an orchard to generate an algorithm that controls the rover on the tasks of localized spraying and fallen fruit collection. Creating and testing these algorithms by using a robotic simulator speed up and ease the evaluation of different scenarios and hypotheses, with the added benefit of being able to test two tasks simultaneously. This method also allows greater freedom and creativity because there are no concerns about hardware damage. It should also be noted that development costs are very low.
- Comparison of on-policy deep reinforcement learning A2C with off-policy DQN in irrigation optimization : a case study at a site in PortugalPublication . Alibabaei, Khadijeh; Gaspar, Pedro Dinis; Assunção, Eduardo; Alirezazadeh, Saeid; Lima, Tânia M.; Soares, V.N.GJ.; Caldeira, J.M.L.P.Precision irrigation and optimization of water use have become essential factors in agricul- ture because water is critical for crop growth. The proper management of an irrigation system should enable the farmer to use water efficiently to increase productivity, reduce production costs, and maxi- mize the return on investment. Efficient water application techniques are essential prerequisites for sustainable agricultural development based on the conservation of water resources and preservation of the environment. In a previous work, an off-policy deep reinforcement learning model, Deep Q-Network, was implemented to optimize irrigation. The performance of the model was tested for tomato crop at a site in Portugal. In this paper, an on-policy model, Advantage Actor–Critic, is implemented to compare irrigation scheduling with Deep Q-Network for the same tomato crop. The results show that the on-policy model Advantage Actor–Critic reduced water consumption by 20% compared to Deep Q-Network with a slight change in the net reward. These models can be developed to be applied to other cultures with high production in Portugal, such as fruit, cereals, and wine, which also have large water requirements.
- Real-time detection of vine trunk for robot localization using deep learning models developed for edge TPU devicesPublication . Alibabaei, Khadijeh; Assunção, Eduardo; Gaspar, Pedro Dinis; Soares, V.N.G.J.; Caldeira, J.M.L.P.The concept of the Internet of Things (IoT) in agriculture is associated with the use of high-tech devices such as robots and sensors that are interconnected to assess or monitor conditions on a particular plot of land and then deploy the various factors of production such as seeds, fertilizer, water, etc., accordingly. Vine trunk detection can help create an accurate map of the vineyard that the agricultural robot can rely on to safely navigate and perform a variety of agricultural tasks such as harvesting, pruning, etc. In this work, the state-of-the-art single-shot multibox detector (SSD) with MobileDet Edge TPU and MobileNet Edge TPU models as the backbone was used to detect the tree trunks in the vineyard. Compared to the SSD with MobileNet-V1, MobileNet-V2, and MobileDet as backbone, the SSD with MobileNet Edge TPU was more accurate in inference on the Raspberrypi, with almost the same inference time on the TPU. The SSD with MobileDet Edge TPU achieved the second-best accurate model. Additionally, this work examines the effects of some features, including the size of the input model, the quantity of training data, and the diversity of the training dataset. Increasing the size of the input model and the training dataset increased the performance of the model.