- Humans and robots work better together following cross-training 2013-02-11
- Spending a day in someone else’s shoes can help us to learn what makes them tick. Now the same approach is being used to develop a better understanding between humans and robots, to enable them to work together as a team.
Robots are increasingly being used in the manufacturing industry to perform tasks that bring them into closer contact with humans. But while a great deal of work is being done to ensure robots and humans can operate safely side-by-side, 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).
“People 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.”
Most existing research into making robots better team players is based on the concept of interactive reward, in which a human trainer gives a positive or negative response each time a robot performs a task.
However, human studies carried out by the military have shown that simply telling people they have done well or badly at a task is a very inefficient method of encouraging them to work well as a team.
So Shah and PhD student Stefanos Nikolaidis began to investigate whether techniques that have been shown to work well in training people could also be applied to mixed teams of humans and robots. One such technique, known as cross-training, sees team members swap roles with each other on given days. “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.
In a paper to be presented at the International Conference on Human-Robot Interaction in Tokyo in March, Shah and Nikolaidis will present the results of experiments they carried out with a mixed group of humans and robots, demonstrating that cross-training is an extremely effective team-building tool.
To allow robots to take part in the cross-training experiments, the pair first had to design a new algorithm to allow the devices to learn from their role-swapping experiences. So they modified existing reinforcement-learning algorithms to allow the robots to take in not only information from positive and negative rewards, but also information gained through demonstration. In this way, by watching their human counterparts switch roles to carry out their work, the robots were able to learn how the humans wanted them to perform the same task.
Each human-robot team then carried out a simulated task in a virtual environment, with half of the teams using the conventional interactive reward approach, and half using the cross-training technique of switching roles halfway through the session. Once the teams had completed this virtual training session, they were asked to carry out the task in the real world, but this time sticking to their own designated roles.
Shah and Nikolaidis found that the period in which human and robot were working at the same time — known as concurrent motion — increased by 71 percent in teams that had taken part in cross-training, compared to the interactive reward teams. They also found that 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 percent.
What’s more, when the pair studied the robots themselves, they found that the learning algorithms recorded a much lower level of uncertainty about what their human teammate was likely to do next — a measure known as the entropy level — if they had been through cross-training.
Finally, when responding to a questionnaire after the experiment, human participants in cross-training were far more likely to say the robot had carried out the task according to their preferences than those in the reward-only group, and reported greater levels of trust in their robotic teammate. “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.”
The work shows that strategies that are successful in improving interaction among humans can often do the same for humans and robots, says Kerstin Dautenhahn, a professor of artificial intelligence at the University of Hertfordshire in the U.K. “People easily attribute human characteristics to a robot and treat it socially, so it is not entirely surprising that this transfer from the human-human domain to the human-robot domain not only made the teamwork more efficient, but also enhanced the experience for the participants, in terms of trusting the robot,” Dautenhahn says.
- Explorations of anticipatory behavioral control (ABC): A report from the cognitive psychology unit of the University of Würzburg 2007-10-30
(2007). Cognitive Processing, 8, 133-142.
- Modeling Top-Down Perception and Analogical Transfer with Single Anticipatory Mechanism. 2007-05-09
Georgi Petkov, Kiril Kiryazov, Maurice Grinberg, Boicho Kokinov (2007); Proceedings of the Second European Cognitive Science Conference, Greece
- Surprise-Driven Belief Revision 2007-04-10
Lorini, E., Castelfranchi, C. (2006); In Proceedings of Second Biennial Conference on Cognitive Science, St. Petersburg, 9-13 June, 2006.
- Establishing Mutual Beliefs by Joint Attention: towards a Formal Model of Public Events. 2007-04-10
Lorini E., Tummolini L., Herzig A. (2005); In Bruno G. Bara, Lawrence Barsalou, & Monica Bucciarelli, Proceedings of the XXVII Annual Conference of the Cognitive Science Society, July 21-23, Stresa, Italy
- Effect of pure error-based fitness in XCS 2007-10-30
(2007). In Kovacs, T. et al. (Eds.) Learning Classifier Systems: International Workshops, IWLCS 2003-2005, LNAI 4399, Springer-Verlag, Berlin Heidelberg. 104-114.
- The Interplay of Analogy-Making with Active Vision and Motor Control in Anticipatory Robots 2007-05-09
Kiril Kiryazov, Georgi Petkov, Maurice Grinberg, Boicho Kokinov, Christian Balkenius (2006); Anticipatory Behavior in Adaptive Learning Systems: From Brains to Individual and Social Behavior, LNAI number 4520
- Anticipatory Behaviour based on Prediction of Image Sequences 2007-06-01
- Simons Center for the Social Brain offering seed grants, postdoc fellowships 2013-01-03
- The Simons Center for the Social Brain (SCSB) at MIT is pleased to announce its 2013 Round 1 funding opportunities for faculty seed grants and postdoctoral fellowships. The deadline for applications is Feb. 28, 2013.
Mission and goals
The mission of the Simons Center for the Social Brain is to understand the neural mechanisms underlying social cognition and behavior, and to translate this knowledge into better diagnosis and treatment of autism spectrum disorders.
Neural correlates of social cognition and behavior exist in diverse species, and the underlying mechanisms will be studied in both humans and relevant model organisms and systems. We expect that experimental approaches will take advantage of strengths at MIT in genetics and genomics, molecular and cell biology, analyses of neural circuits and systems, cognitive psychology, mathematics and engineering.
MIT faculty members with an interest in autism research may apply as the PI on a seed research grant. We seek innovative research projects that are directly relevant to autism and that bridge at least two different MIT labs, or one MIT lab and another at a Boston-area institution (typically a hospital). The expectation is that the seed funds will enable the collection of pilot data on bold new projects, bringing the work to the point where it can be funded through standard channels after the first year. This mechanism will provide a single year of support, at a maximum level of $100,000 in direct costs (indirect costs need not be included in the budget).
The project must involve one MIT faculty member as PI and at least one other independent researcher as co-PI from a different lab. When a MIT PI applies with a co-PI from another Boston-area institution, the funds will be budgeted for spending at MIT. Successful applicants can apply later for a second year of funding, but the application will be considered in competition with all submitted applications (including new ones).
Applications for postdoctoral fellowships (named Simons Postdoctoral Fellowships) are sought from candidates with PhD or MD degrees who aim to conduct research at MIT that is relevant to autism. These prestigious fellowships are open to candidates nationwide. They are designed to enhance and showcase autism research at MIT and will be awarded to candidates who propose innovative research bridging at least two different labs.
Each postdoctoral applicant must have a primary advisor who is a MIT faculty member, and a secondary advisor who is an independent researcher at another MIT lab or Boston-area institution. While the fellowships are open to candidates currently at MIT, our goal is to attract outstanding external candidates. MIT faculty members are encouraged to bring these fellowships to the attention of exceptional candidates who wish to come to MIT for postdoctoral training as Simons Fellows.
The Simons Postdoctoral Fellowships will provide a competitive stipend plus an allowance for health insurance, travel and research-related expenses. The fellowships will be awarded for 2 years, conditional upon satisfactory progress at the end of the first year.
For information on how to apply for either seed grants or postdoctoral fellowships, please visit the SCSB website.