They say, to understand somebody try spending a day in his shoes. A team of MIT scientists is trying this credo to bring about better cooperation between robots and humans.
“More effort is needed to make robots smart enough to work effectively with people,” says Julie Shah, an assistant professor of aeronautics and astronautics at MIT and head of the Interactive Robotics Group in the Computer Science and Artificial Intelligence Laboratory [CSAIL].
“Humans aren’t robots, they don’t do things the same way every single time,” Shah says. “And so there is a mismatch between the way we program robots to perform tasks in exactly the same way each time and what we need them to do if they are going to work in concert with people.”
Current research helps robots understand humans through a process of feedback in which the human trainer gives a positive or negative response each time a robot performs a task. However, studies carried out by the military have revealed that just telling people they have done well or badly at a task is a very inefficient way of encouraging them to work well as a team.
Therefore Shah and PhD student Stefanos Nikolaidis began their experiments to explore the feasibility of applying techniques that work well teams composed of humans and robots. One such technique, known as cross-training, makes team members swap roles with each other. “This allows people to form a better idea of how their role affects their partner and how their partner’s role affects them,” Shah says.
Shah and Nikolaidis will present the results in March 2013 demonstrating that cross-training is an extremely useful technique to make a human and robot effective team-mates.
To enable robots to participate in the cross-training experiments, the pair had to change the programming in the robots. They modified existing reinforcement-learning algorithms to allow the robots to take in not only information from positive and negative rewards, but also obtain information through demonstration. In this manner, by watching human carry out their tasks, the robots were able to understand how humans wanted them to perform the task.
Armed with skill, a robot was teamed up with a human and they performed a simulated task in a virtual environment. Half of the teams used the conventional interactive reward approach, and half the teams used a cross-training technique of switching roles halfway through the session. Once the teams had completed this virtual training session, they the required do the task in the real world, but without switching this time.
The team found that the period in which human and robot were working at the same time—known as concurrent motion—increased by 71 per cent in teams that had taken part in cross-training, compared to the interactive reward teams. This means both of them were being efficient with lesser idle time spent in one observing the other. The amount of time the humans spent doing nothing—while waiting for the robot to complete a stage of the task, for example—decreased by 41 per cent.
The researchers found that the learning algorithms in the robots reported a highly reduced level of uncertainty about what their human team-mate was likely to do next if the robots had been through cross-training.
Also, many human participants in cross-training felt that the robot had carried out the task according to their preferences unlike those in the reward-only group who were fewer. Humans also trusted their robots more after cross-training. “This is the first evidence that human-robot teamwork is improved when a human and robot train together by switching roles, in a manner similar to effective human team training practices,” Nikolaidis says.
Shah believes this improvement in team performance could be due to the greater involvement of both parties in the cross-training process. “When the person trains the robot through reward it is one-way: The person says ‘good robot’ or the person says ‘bad robot,’ and it’s a very one-way passage of information,” Shah says. “But when you switch roles the person is better able to adapt to the robot’s capabilities and learn what it is likely to do, and so we think that it is adaptation on the person’s side that results in a better team performance.”
If robots can learn to cooperate with other humans, may be it is time we learnt to cooperate with each other too. After all, we are all human.