This paper explores the integration of quadruped robots in extensive livestock farming as a novel application of field robotics. The SELF-AIR project exemplifies this innovative approach, leveraging quadruped robots with advanced sensors, artificial intelligence, and autonomous navigation systems to assist farmers in various tasks such as herd monitoring, health assessment, and environmental management.
The paper provides an overview of the SELF-AIR project, outlining the key requirements for a shepherd robot, including perception, navigation, path planning, obstacle avoidance, interaction with sheep, communication, endurance, and adaptability to environmental conditions. It also introduces the hardware platforms used, the Unitree A1 and Ghost Vision 60 quadruped robots.
The paper then presents the findings from a series of field tests conducted on different-sized farms, ranging from a small private farm with 2 sheep to a medium-sized research farm with 100 sheep. These experiments explored the dynamics of herding using the quadruped robots, both individually and in combination with drones, as well as the natural herding behaviors observed without the use of robots.
The lessons learned from these deployments are discussed, highlighting the advantages and disadvantages of integrating quadruped robots in livestock farming. Advantages include enhanced livestock monitoring, efficient pasture navigation, extended patrols, effective herding, behavioral insights, all-weather monitoring, remote herding control, and increased overall efficiency. Disadvantages include challenges with animal interaction, flock disturbance, limited endurance, herding instinct, data privacy, training requirements, equipment protection, signal range, resistance to change, regulatory restrictions, and cultural resistance.
The paper concludes by emphasizing the significant potential of quadruped robots in livestock farming and the need for further research and development to refine their autonomous capabilities and optimize their performance in real-world farming environments.
To Another Language
from source content
arxiv.org
Deeper Inquiries