Robots could scour avalanche sites, enter burning buildings or secure city streets contaminated by poisonous chemicals, saving lives and increasing the effectiveness of rescue missions.
Scientific advances in robotics research are moving the technology from predictable spaces like production lines into disaster zones. After earthquakes, accidents, avalanches or explosions, robots can take the place of their human teammates, cutting risk to human life and helping boost the chances of rescuing victims.
Cutting-edge research projects like NIFTi, WALK-MAN and SHERPA are changing the way rescue services approach disaster situations, and improving the prospects for everyone involved – victims and rescuers alike.
The EU-funded NIFTi project is using a ground robot and a robotic helicopter to help out in urban rescue scenarios.
‘We wanted to take the technology from a structured indoor environment into the real world, where it is less structured and there are things you don’t expect – that makes all dimensions much more difficult,’ said coordinator Dr Ivana Kruijff-Korbayová, of the German Research Center for Artificial Intelligence.
The researchers started off thinking of the robot as a simple tool to be used by rescuers, but it ended up with a more expanded role within the team, Dr Kruijff-Korbayová said.
‘As the project developed, it turned out that the collaboration between humans and robots needs to go much further – rather than seeing robots as tools or mobile cameras, they need to be seen as a team member.’
‘Rather than seeing robots as tools or mobile cameras, they need to be seen as a team member.’
Dr Ivana Kruijff-Korbayová, coordinator of the NIFTi project
The project included the Dortmund Fire Department and the Italian fire brigade forces, which allowed researchers to engage end users and see how the robot would work in real emergency situations.
‘They could not use a tool if it would get in the way and cost lives,’ Dr Kruijff-Korbayová said. ‘However, in a situation where there is a danger – for example of radioactivity or chemical leak or an accident in a tunnel where you don’t know how hot it is – if you would be putting a firefighter at risk, the usefulness of this technology already in its current state becomes more obvious.’
To be most useful in a disaster situation, robots need to work hand in hand with humans. In the case of an avalanche, robots could scour the skies and the hillside, leaving a human rescuer to think strategically.
That’s the idea behind the EU-funded SHERPA project, in which two aerial vehicles and a ground rover work to support the human rescuer, who is known as the ‘busy genius’.
‘The busy genius has strong experience in the field – even the best robots cannot match their experience or cognitive ability – but emotion plays a crucial role so sometimes the vehicles cannot expect input,’ said Professor Lorenzo Marconi at the University of Bologna, Italy, who coordinates the project. ‘From an academic point of view, it’s interesting to have a complicated team. The human in the loop is quite a big issue.’
Crucially, the technology must not get in the way of efforts by the human rescuer to locate a missing person. So commands are given not with a joystick, but through gesture and voice, with the help of electronic glasses. ‘You cannot ask the operator to waste time – it must be as autonomous as possible without flooding the person with huge amounts of data,’ said Professor Marconi.
NIFTi robot on a reconnaissance mission in the Duomo in Mirandola, Italy, in the Emilia-Romagna region mission July 2012. © NIFTi Team
Dr Kruijff-Korbayová of the NIFTi project believes wider use of robotic technologies in disaster situations will come in the near future, with some ground robots already in use and helicopters not far off.
There are some hurdles still – money, awareness and development of the technology so that it can easily be used by non-experts – but they are not insurmountable. ‘What seems to have been learnt from NIFTi is that it is the whole package – visual processing, autonomous ability to drive through terrain and so on – that is so important,’ she said.
NIFTi took part in the assessment of the situation after the earthquake that struck Mirandola in Emilia-Romagna, northern Italy, in 2012.
‘One big lesson we learnt from that was the importance of persistence,’ Dr Kruijff-Korbayová said. So, following on from the NIFTi project comes the TRADR project, where the key word is persistence: ‘The idea is to gather information over a longer time and learn from it so that it is not just information but experience.’
That paves the way for greater autonomy – or at least partial autonomy – for the robot, which can learn from experience and not rely solely on the input of its operator. For that, the operator’s trust is key and the TRADR project will look at improving interaction and communication – here too, keeping the human in the loop is vital.
Another EU-funded project – WALK-MAN – is developing a human-like robot to operate in realistic places such as buildings that have been damaged by natural catastrophes or man-made disasters.
‘Recent events showed the necessity for reliable and effective robotic systems that can be deployed rapidly after the disaster, to assist in tasks too hazardous for humans to perform,’ said Dr Nikos Tsagarakis, scientific coordinator of the project.
The robot will have powerful manipulation skills to be able to lift up collapsed masonry, walk and crawl over obstacles and through cluttered spaces, and be sturdy enough to operate tools like drills or cutters. It will be able to work autonomously or by remote control if communications are limited or unreliable.
To achieve these objectives, the WALK-MAN project will develop technology that allows for whole-body locomotion and manipulation - meaning all of the robot’s body parts help ensure stable and balanced movement.