In the middle of one of the halls is also the station of the Faculty of Czech Technical University in Prague.
The robot was created, among other things, thanks to participation in a prestigious robotics competition in the United States, where scientists developed algorithms for rescue teams. The robot's main task is to find the injured.
Robotic 'dog' from Czech Technical University in action
"Hello, I'm MARV and I'm exploring this area. My task is to find the injured. Unfortunately, rescuers can't come here right now due to the conditions. That's why I'm here. My operator is listening. Is anyone injured and needs help? If you can walk, follow me. I'll lead you to safety," says the robot.
Exploration robots are intended primarily for areas dangerous to humans. “When the environment becomes risky – for example, after an earthquake or when a building partially collapses – it is dangerous to send a large number of firefighters there, who would have to work there for a long time. That is why both firefighters and reconnaissance robots are deployed to such places. They map the area, search for victims and provide the firefighters with information so that they can go directly to the scene and rescue people,” describes Číhala.
This is also advantageous in places contaminated with toxic gas. Firefighters there have to work in breathing apparatus, which is very demanding for them, so robots are ideal for such use.
The goal is to teach robots to “see” as well as possible in difficult conditions and to be as independent as possible. And this is precisely where the field has made significant progress in recent years.
“We participated in a prestigious competition in America, where we drove in a structured environment – factories, rooms, doors or stairs. This is already quite well known from a scientific point of view and robots can navigate well in this autonomously. We are now focusing more on driving in nature, recognizing passable grass and the ability to overcome stones, logs, enter a stream and generally orient ourselves in nature,” explains Číhala.
Thanks to cameras and sensors, Marv sees better than a person. “Controlling a robot in a crisis environment is not easy for an operator, but a robot really sees more than a person. The robot sees in the infrared spectrum, so it can recognize heat, and thanks to a laser 3D scanner it knows what the area around it looks like. The cameras allow it to look in all directions.”
Easier control, greater autonomy
The idea of the future of these technologies – that robots will be launched into the field and humans will only monitor the situation from a distance – is tempting. According to Číhala, however, we will still have to wait for that.
“I believe that it will happen one day, but I don’t think it will happen in the next five or ten years. Right now, we are aiming for the decision-making element of people and their experience to always be there.”
The human factor will always play a role in controlling robots. However, the aim of the research is to simplify this method of control as much as possible. “We want humans to assign tasks to robots rather than having to control their every movement. The ideal situation would be one where the operator only monitors the data that the robot collects and evaluates it, while the movement in the field itself can be handled by the machine,” adds Číhala.