Sensory Artificial Intelligence Can Robots Have Senses Like Humans?
The great advances in AI we are witnessing today are driven by recent advances in deep learning , which allows machines to learn automatically during training. This eliminates the need for the tedious work of manually adding new features, which is one of the main reasons why deep learning is a pivotal force in technological innovation and AI.
Building on this development, Meta, the owner of Facebook, WhatsApp and Instagram, has entered a new field by developing tactile AI technologies. The company recently introduced three new tools based on this type of AI: Sparsh , Digit 360 and Digit Plexus to give robots a touch sensitivity that closely mimics human perception.
But what is the purpose of developing these tools?
The ultimate goal is to create robots that can not only mimic tasks, but actively interact with their environment, just as humans interact with the world.
Sparsh is a general AI model that allows robots to interpret and react to sensory signals in real time, while Digit 360 is a prosthetic fingertip for robots that includes a sensor that can perceive touch and subtle physical sensations, such as pinpricks.
Digit Plexus provides a unified framework for integrating tactile sensors across robots of different designs, making it easier to collect and analyze tactile data.
Meta believes these tools will enable robots to perform complex tasks that require a “human touch,” especially in areas like healthcare, where precision and sensitivity are essential.
Impact of sensory robots on society:
The development and use of sensory robots in various fields raises many questions, most notably: Can this technology open up new levels of cooperation, or will it introduce complexities that society may not be ready to deal with?
“As robots gain new senses and greater intelligence and autonomy, we will have to start thinking about their role in society,” says Ali Ahmed, co-founder and CEO of Robomart . “Meta’s efforts are a major first step toward equipping robots with human-like senses. Over time, humans may begin to treat robots as companions, perhaps building a life together.”
In addition to developing sensory AI, Meta also unveiled the PARTNR benchmark, a unified standard for evaluating human-robot collaboration capabilities at scale. Designed to test interactions that require planning, reasoning, and collaborative execution, the benchmark will enable robots to navigate environments alongside humans.
PARTNR leverages large language models (LLMs) to guide these interactions, making bots true partners capable of working seamlessly with humans.
However, some experts say challenges remain. “There will be a lot of work needed to scale this research up to practical applications in real-world environments,” says Ram Palaniappan, CTO of TEKsystems .
Meta Partners to Bring Sensory Robots to Market:
Meta is partnering with GelSight Inc. and Wonik Robotics to develop and commercialize these innovations. GelSight will produce the Digit 360 sensor, which is due to launch next year, and Wonik Robotics will produce the next-generation Allegro robotic hand, which supports Digit Plexus to perform delicate tasks with precision.
But not everyone is convinced that these developments are a step in the right direction. “While adding sensing capabilities can be useful for robots to understand the environment, I don’t think we’re anywhere near giving them human-like sensations, and I don’t think that’s really necessary,” says Agustin Huerta, vice president of digital innovation at Globant .
Are we ready to deal with robots that feel?
As the robotics industry continues to advance, an important question arises: Are we ready for a future where robots can “feel”? And will this technology reshape our relationship with machines?
Some experts believe that applying sensory AI could be a crucial step towards robots understanding the physical world better than ever before, paving the way for greater collaboration between humans and machines.
Other experts say this revolution may not be different from the industrial revolution, but it will require educating future generations to develop a healthy relationship between humans and robots.