Without a doubt technology evolves by leaps and bounds, and with the aim of solving cognitive tasks, Hyperdimensional Computing proposes to calculate the similarity between the data which allows a fast learning ability, energy efficiency and an acceptable precision in the learning and classification of tasks. On the other hand, it also helps data transformation with an intrinsically robust nature.
Hyperdimensional computing uses concepts of hypervectors where these have versatile properties that are applicable to the functioning of Artificial Intelligence.
Hyperdimensional computing and robotics
For robots to be capable and intelligent like humans in various tasks, they need to coordinate sensory data with the capabilities of the robotic engine. Scientists at the University of Maryland published an article in their journal Science Robotics describing a potentially revolutionary approach to improving the way Artificial Intelligence handles sensory-motor representation using hyperdimensional computing theory.
The researchers aimed to create a way to improve a robot's "active perception" and the robot's ability to integrate the way a machine will fit into the world around it. So, they proposed a method to codify actions and perceptions together in a single space that is meaningful, semantically informed and consistent through the use of this technology.
Using these vectors, researchers can keep all the sensory information the robot receives in one place, so with this, the robots could keep their memories. As more information is stored, your history will increase the memory content of the machine. This will result in robots being better at making autonomous decisions, waiting for future situations, and completing tasks.