Meta on Friday announced PARTNR, a new program designed to study human-robot interaction (HRI). The research is specifically focused on how humans and robots might collaborate in the home environment. That includes mundane tasks like cleaning, cooking, and picking up food deliveries.
Automated housework is a decades-old dream, most prominently captured by Rosey on “The Jetsons.” The robot maid debuted on prime-time television more than 60 years ago but continues to be a popular touchstone when discussing the potential for sophisticated machines to remove some of the burden of household chores.
To date, however, only the robot vacuum has made significant headway in the market. There are numerous reasons why no other home robots have cracked the mainstream, including price, reliability, and limited functionality. It’s certainly not for lack of trying, nor is it due to consumer disinterest. It’s just that no other robot has hit the mark on both cost and feature set.
It’s likely that seeing more robots in the home will require improved collaboration with the people who own them. The first wave of home robots are unlikely to single-handedly manage chores. Even a good robot vacuum needs help from time to time. Meta is positioning PARTNR as both a benchmark and dataset to determine how people and robots might work together to get things done around the home.
“Our benchmark consists of 100,000 tasks, including household chores such as cleaning up dishes and toys,” Meta writes. “We are also releasing the PARTNR dataset consisting of human demonstrations of the PARTNR tasks in simulation, which can be used for training embodied AI models.”
Simulation has become an increasingly useful tool in robot deployment, allowing organizations to test in seconds what might otherwise take hours or days to accomplish in the real world. Meta says, however, that it has also had success deploying the PARTNR model outside of simulation. It has already been used in Boston Dynamics’ Spot robot in testing. Meta has also built a mixed-reality interface designed to offer a visual representation of the robot’s decision-making processes.
“The potential for innovation and development in the field of human-robot collaboration is vast,” Meta adds. “With PARTNR, we want to reimagine robots as future partners, and not just agents, and jump start research in this exciting field.”
Age tech holds a lot of potential for the category. Labrador’s automated serving cart, for example, offers insight into ways technology might assist older people who continue to live independently. However, many advances of the variety Meta is aiming to address will be required before such systems gain mainstream acceptance.
Humanoids are another intriguing avenue that have presented themselves in recent years. Most companies behind these bipedal robots foresee a future in which they will eventually help out in the home. That said, pricing needs to come down considerably and reliability needs to improve by leaps and bounds. That is a large part of the reason most manufacturers are looking to address corporate needs first.
With the right scaling and advancements in AI, one can imagine a world in which humanoid robots address general-purpose tasks in a way that allows them to help in both the factory and the home. A major stepping stone to that place requires solid advancements in human-robot collaboration. Meta, which has been exploring robotics amid its wider AI research, is hoping that PARTNR can help them get there.
Published by: Brian Heater
