On Aquaculture Enhancements Through Robotic Behaviour Actuation

• Letieri Rodrigues de Ávila FURG
• Sílvia Silva Da Costa Botelho FURG
• Marcelo Rita Pias FURG
• Bruna de Vargas Guterres FURG
• Je Nam Jun Junior FURG

Resumo

Shrimp farming activities in Brazil have increased tremendously to complement natural fishing. It motivates the applications aimed at controlling and managing shrimp farming systems. Computational and automation based techniques provide the ground for a new generation of smart systems. To this end, computer vision systems can improve cultivation processes by supporting biomass generation and animal health monitoring. Shrimp’s ecological niche is one of the most significant computer-vision obstacles in managing such systems due to turbid watery habitat, nocturnal foraging, and rapid displacement. The magnetotactic behaviour of shrimps becomes an alternative for "guiding" the animals towards the field of view of a camera. The present work proposes and discusses an Robotic Behaviour Actuator System (RBAS) for biomass estimation based on computer-vision techniques and behavioural control of Litopenaeus vannamei. The proposed system addresses layout design to consider animal behavioural reactions and welfare in long-term sustainable aquaculture.

Palavras-chave: Animals, Robot vision systems, Estimation, Lighting, Cameras, Real-time systems, Biomass