How horses are lighting up the future of robot-human interaction

Eakta Jain learned to ride a horse in order to understand how horses can influence the design of future robots. UF News.

One might think that a future where animals are replaced by robots is dystopian. However, whether it is horses carrying or pulling heavy loads or guard dogs, the animals at work have already been supplanted by machines. Think of automobiles or IoT-based alarm systems.

And today we are at the dawn of a new change. Another type of machine will perform even more complex tasks for humans, especially in the fields of health, education and industry.

Yet, at present, there are no “guiding principles” governing the interaction between humans and robots. In other words, we need to think about how humans and robots should communicate with each other to maximize efficiency and minimize disruption – and maybe even build bonds?

The science fiction writer Isaac Asimov was the first to try to frame this relationship when he created, in 1942, a set of rules. These rules are as follows: don’t let human beings harm you, obey orders (and don’t break rule 1) and protect yourself (and don’t break rule 2).

Why dogs are not necessarily the right example

At a time when we are talking about the disappearance of humanity due to the impact of AI, Asimov’s rules may seem sensible, even if they are a little vague. Unfortunately, these rules are also too simplistic and do not really inform about the complexity of current robots.

The study of the behavior of dogs and the way in which they influence the relationship with their masters proposes another way of studying the thing. Because the canids seem to appreciate human companionship as well as leadership. The problem, however, is that dogs are predators and that they are not the best models for the design of all robot machines. They also accompany most of their behaviors with sounds, such as barking, growling and moaning.

This is the kind of problems that Eakta Jain, associate professor of computer science at the University of Florida, faced when she tried to conceptualize a framework for human-robot interactions. She was able to observe how autonomous vehicles are able to monitor other vehicles and maintain an appropriate distance from each of them, and even monitor the driver.

“Horses have been interacting with humans for over 10,000 years”

It was while trying to define a reference to develop principles for the design of robot-human interaction that she had an idea. “Horses have been interacting with humans for more than 10,000 years, for the transport of goods and people, farm work and wars. When I started thinking about this analogy, I realized that there were many parallels with what robots will do,” says Ms. Jain in an interview with ZDNET.

In fact, horses are extraordinary beings. They have an amazing ability to perceive their surroundings and are very receptive to their master. They are also extremely intelligent and can be trained to work as a team.

Above all, horses have been used all over the world, in different cultures and geographical areas. “I also realized that there was something in them that resembled a universal interface with humans,” says Ms. Jain. To better understand the behavior of horses, she spent time in an equine center. And she realized that the only way to get a real idea of a horse’s behavior was not only to observe, but also to learn how to ride a horse.

Horses communicate mainly through non-verbal methods

For six months, Mrs. Jain learned to ride a horse through weekly lessons. And she was able to identify some key principles of observation, which could be used for the design of robots.

The first and most important of these principles is that horses communicate mainly through non-verbal methods. They move their ears to follow the sound of movements and threats. The horses also move their ears to indicate whether they are listening to you or not. Other non-verbal signs include the accumulation of tension in the muzzle, the raising of the neckline, or the trepining.

So, how to apply this knowledge in a robotic future? The researchers imagined a scenario in which non-verbal signals could play an important role, for example in the case of a person suffering from a hearing impairment.

Horses trained at the University of Florida, Horse Training Unit in Gainesville. University of Florida

A therapeutic robot with prominent ears

A therapeutic robot could, for example, have prominent ears that point towards the human user when he listens to him, and towards the door when he hears knocking.

The notion of respect is another aspect that Ms. Jain and her colleague found peculiar to horses. “We don’t think about respect in the context of human-robot interactions,” said Ms. Jain. “In what way can a robot show you that it respects you? Can we conceive of behaviors similar to those of the horse? Will this make humans more inclined to work with the robot?”

Young horses are trained early on to move away or back away when a human approaches them. Similarly, a horse that moves at the pace of the trainer indicates that he holds the human in high esteem.

Configure the direction of the hierarchy

Hospitals, where robots should help nurses and doctors, following them, stopping and restarting in tandem with their human counterparts, are also ideal spaces to illustrate this design principle in a real scenario.

If you remember the quarrels between R2-D2 and C-3PO in The Star Wars, for example, you may have an idea of the confrontation that could take place between different types of robots, which refuse to give way to each other. By using the behavior of the horses, such as size or type to determine the hierarchy of dominance and, consequently, respect, the robots can be designed to give way to each other according to their position in the established hierarchy, and to determine who has priority to deliver his item first in the context of delivery robots.

The research of Mrs. Jain also show that different horses have different abilities. Why? Because they are bred for specific characteristics. Likewise, robots and humans have different abilities and, when working together, different speeds. Thus, any future relationship between a human and a robot requires that both parties get used to each other and learn as they work together during the first stages of the relationship.

Learning to walk side by side with a human

For example, a packet-carrying robot that must stand at a determined distance from a human must learn to walk at variable speeds. And this is so that the robot can acclimatize to the walking pattern of this specific human. And this by taking into account the speed, pace or rhythm. In other words, the robot will adapt its habits to the man with whom it works.

This type of early learning through repetitive tasks is particularly crucial in certain scenarios, such as in a factory, where the man has a particular posture or gait. This educational process will make it possible to avoid slip-ups or the development of a counterproductive feeling of disdain towards the robotic partner.

At the dawn of a new era for robots, the way we communicate with our counterparts, the methods with which we train them and the way we consider them could become a crucial marker of success.

Source: “ZDNet.com “