Smart cows: The University of Sydney’s digital cow “Swagbot” has received a substantial upgrade, entering the Artificial Intelligence (AI) era. Once a relatively simple robot with limited functions, such as walking through difficult terrain, the digital cow now boasts sensors, AI and self-learning mechanisms. The objective of the scientists is to create the world’s first “smart cow,” helping to make livestock farming more efficient and climate friendly.

The four-wheeled autonomous robot can work 24 hours a day, reducing the need for fossil fuels by running on electricity and solar power. 

Moreover, the digital cow can monitor the health of the livestock, as well as determine the type and density of the pasture it works on. After gathering the information, the machine uses it to autonomously herd the cattle to the best pastures and move them before the land is degraded and overgrazed.

Development in the field is urgently needed as the leader of the smart cow project, Professor Salah Sukkariew, believes that agriculture and farming are facing demand for greater productivity due to a global population surge and rising incomes in developing countries. By 2050, experts predict that global food demand will increase by between 59% and 98%.

Virtual cows herding helps us understand AI

In a separate study, researchers from Macquarie University in Australia, Scuola Superiore Meridionale, the University of Naples Federico II, and the University of Bologna in Italy, and University College London in the UK used a videogame where players herded cattle to understand how humans make navigational decisions (how we coordinate our movements in response to what is happening around us).

The study will aid the understanding of dynamical perceptual-motor primitives (DPMPs), which can be used for AI to mimic human decision making. “It could help us not only interact more effectively with artificial intelligence, but even improve the way robots move in the future,” the researchers explain.

How do we actually map our surroundings?

Scientists have unveiled that our previous assumptions on how we mapped our surroundings were probably wrong.

Previously, researchers assumed that our brains were rapidly making detailed maps of our surroundings, then planning how to move through them. However, the scientists explain that “an increasing body of research” now supports the idea that rather than making a detailed plan, we move naturally, taking into account our goal and making allowances for any obstacles we encounter along the way.

To find this, participants were asked to work on two herding tasks in a videogame, moving either a single cow or a group of cows into a pen. Researchers found that players chose the cows closer to them in the angular distance, and when choosing between two cows, they most likely chose the ones that were furthest from the centre of the containment zone.

After providing the AI with the decision-making rules found in participants, it could predict “nearly 80% of choices” of which cows to herd next.

In other “cow news,” earlier this year Japanese space startup Interstellar Technologies developed a method to transform the methane in cow waste into rocket fuel.