Robots given ‘common sense’ to help navigate the world
Image credit: Dreamstime
Carnegie Mellon University researchers have developed a system for robots to better understand their environment by recognising the context of different objects.
A robot travelling from point A to point B is more efficient if it understands that point A is the sitting room sofa and point B is a refrigerator, even if it is in an unfamiliar place. This is the basis for the “semantic” navigation system which has been dubbed SemExp.
The system was developed, using machine learning to train a robot to recognise objects, such as understanding the difference between a kitchen table and an end table with knowledge about where in a home such objects are likely to be found.
This enables the system to think strategically about how to search for something, said researcher Devendra Chaplot.
“Common sense says that if you’re looking for a refrigerator, you’d better go to the kitchen,” Chaplot said.
Classical robotic navigation systems, by contrast, explore a space by building a map showing obstacles. The robot eventually gets to where it needs to go, but the route can be circuitous.
Previous attempts to use machine learning to train semantic navigation systems have been hampered because they tend to memorise objects and their locations in specific environments. Not only are these environments complex, but the system often has difficulty generalising what it has learned to different environments.
Chaplot sidestepped that problem by making SemExp a modular system with semantic insights allowing it to determine the best places to look for a specific object: “Once you decide where to go, you can just use classical planning to get you there,” he said.
This modular approach turns out to be efficient in several ways. The learning process can concentrate on relationships between objects and room layouts, rather than also learning route planning. The semantic reasoning determines the most efficient search strategy. Finally, classical navigation planning gets the robot where it needs to go as quickly as possible.
Semantic navigation ultimately will make it easier for people to interact with robots, enabling them to simply tell the robot to fetch an item in a particular place, or give it directions such as “go to the second door on the left”.
Sign up to the E&T News e-mail to get great stories like this delivered to your inbox every day.