Large volumes of data imply a true revolution for robotics: machine learning multiplies the response capacity to the environment, making robots more autonomous, intelligent and agile.

Compared to other fields, robotics has particular characteristics, because it aims to allow a physical agent to interact with the concrete world. 

Traditionally, the field of robotics has been dominated by the desire to find solid models based on the response principle: Optimize the relationship between the input of a stimulus and the output of the action or the prediction of a response to a given stimulus already stipulated.

This worked well in a factory world with well-defined inputs and outputs. However, for years, it has been a challenge for robots Leaving perfectly structured factory or laboratory environments into the “real world”, characterized by its wide range of possible tasks, stimuli and situations.

Machines are advancing rapidly in the field of stimuli and perception: they are beginning to see, hear, read, and touch in ways that were not possible before. This has meant a real revolution for robotics: the ability to learn and deal with direct inputs from the real world greatly enriches the capabilities of robots. 

And here, the large volumes of information are key. This is so because the possibilities of machine learning multiply. The availability of Big Data and associated new learning techniques have the potential to allow robots to understand and operate in significantly more complex environments. Naturally this should lead to a quantum leap in the performance and implementability of robotics in a wide range of practical applications and real environments.

Arthur Dubrawski, director of the Auton Lab – the robotics institute at Carnegie Mellon University's School of Computer Science – explains that robots have always been based on data, because the operational definition of the robot is about doing the following sequence in a loop: detect, plan and act. 

What happens is that Artificial Intelligence and Machine Learning will be more and more protagonists of this process.

 Looking to the future, robotics and manufacturing they will be increasingly defined by big data and the ability of AI systems to analyze and act on production information.

robots now can acquire skills to perform tasks through advanced machine learning, enabled by the development of embedded sensors and cloud-based analytics. This trend will continue, aided by faster 5G wireless networks.

Factories of the future will use Big Data collection and analysis to enable robots make quick decisions in the manufacturing process, even when presented with unfamiliar equipment and objects. The search will be towards robots more autonomous, intelligent, receptive and agile.

 Julio Cesar Blanco – November 14, 2022