Artificial intelligence (AI) that reasons like a human remains elusive, but Honda hopes to make inroads. The Tokyo company’s U.S.-based Research Institute today announced a collaboration with three academic institutions — the Massachusetts Institute of Technology (MIT), the University of Pennsylvania (Penn), and the University of Washington — to advance the field of artificial cognition.
MIT’s Computer Science and Artificial Intelligence (CSAIL) lab, in partnership with Penn’s School of Engineering and Applied Science and the University of Washington’s Paul G. Allen School of Computer Science & Engineering, will develop prototypes, working examples, and demonstrations of what Honda calls the “mechanisms of curiosity.” Specifically, MIT CSAIL will focus its efforts on systems capable of predicting future percepts — concepts developed as a consequence of perception — and the effect of future actions, while Penn’s engineering department and the Paul G. Allen School will develop perception models informed by biology and robots that can work safely in human environments.
Grants will fund the first leg of research. According to the University of Washington’s newsroom, the Paul G. Allen School-led team will receive $2.7 million for a social robot that interacts with people in a building or home, and a robot arm that can manipulate objects placed in front of it.
“Our ultimate goal is to create new types of machines that can acquire an interest in learning and knowledge, and the ability to interact with the world and others,” Soshi Iba, a principal scientist at Honda Research Institute, said. “We want to develop Curious Minded Machines that use curiosity to serve the common good by understanding people’s needs, empowering human capability, and ultimately addressing complex societal issues.”
The Curious Minded Machine initiative expands on Honda’s earlier forays into so-called cooperative machines — i.e., robots that not only work effectively together, but that operate transparently and in a way that builds confidence and trust. At the 2018 Consumer Electronics Show in January, it announced a new lineup of social and assistance robots under the moniker 3E (short for Empower, Experience, and Empathy), including an actuated barstool, a motorized camping cooler, and an autonomous four-wheeler.
Honda’s pivot toward assistive robotics can be traced back to July 2013, in fact, when it unveiled a disaster response robot based on Asimo. Four years later, it debuted the E2-DR, a disaster response robot that can climb stairs and stepladders, perform environment planning and monitoring, and avoid obstacles and debris.
Not all of its experiments projects have borne fruit. In June, Honda canceled development of its iconic Asimo humanoid robot, the newest version of which — announced in 2011 — could run, climb stairs (albeit slowly), and even dab. For a good portion of the three decades following its 1986 debut, it arguably lived up to the slogan of “the world’s most advanced humanoid robot” — at least until Boston Dynamics’ ultra-dexterous Spot and Atlas came along.
Honda joins the Defense Advanced Research Projects Agency (DARPA), a division of the U.S. Department of Defense, in the pursuit of AI capable of thinking independently. The agency is investigating computational models that mimic core domains of cognition — objects (intuitive physics), places (spatial navigation), and agents (intentional actors) — as part of its Machine Common Sense Program.
According to a survey of AI researchers published at the Joint Multi-Conference on Human-Level Artificial Intelligence (HLAI) held last month in Prague, artificial general intelligence — machines that can perform any task a human can — could emerge in the next 10 years.
But others are less optimistic. In an editorial investigating Facebook’s work in false news-flagging AI, two New York University professors — Gary Marcus, a professor of psychology and neural science, and Ernest Davis, a professor of computer science — argued that short of a paradigm shift, truly human-like AI is beyond reach.
“To get to where Mr. Zuckerberg wants to go will require the development of a fundamentally new AI paradigm, one in which the goal is not to detect statistical trends but to uncover ideas and the relations between them,” they wrote. “Only then will such promises about AI become reality, rather than science fiction.”