Advanced sheep posture controller

• Autor(a) principal: Cardoso, António Ferreira
• Data de Publicação: 2018
• Tipo de documento: Dissertação
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10773/25599

Weed control in vineyards is an actual and important problem in modern viticulture. Traditional methods, mechanical- or chemical-based, are expensive and hazardous to the environment and cultures. Controlling weed infestation through pasturing animals is an ancient and environmentally friendly technique. However, animals cannot be used on vineyards for the entire production cycle, because they tend to feed from vine branches and fruits. The SheepIT project aims to solve this limitation through automated behavior monitorization and conditioning of sheep pasturing in vineyards. To achieve this, an IoT network architecture, with electronic collars placed on the animals' neck at its base, was designed and implemented, prior to this work's development. This dissertation describes the design, implementation and validation of the posture control algorithm, whose objective is to teach sheep, through stimulation, to eat only ground weeds, avoiding vine branches and grapes. The algorithm relies on sensors installed in the collar to retrieve information on the sheep's posture, and conditions its behavior through actuators, also embodied on the collar. These activities are carried out concurrently with other system services, e.g. virtual fencing and flock data logging, which are out of the scope of this dissertation.