Science fiction and animation were an enabling environment to reflect the fascination humans feel towards robots. Astro Boy, Mazinger Z, Wall-E, Bumblebee, and even Terminator are just a few examples.
In recent years, we’ve seen robots emerge that provide multiple jobs, from catering, preparing food, cleaning hospitals to guarding parks. But despite its extraordinary usefulness, it is still far from possessing the abilities we have seen in literature, movies and manga. However, this may change very soon. Recently, a major advance in computer science has been revealed that although it will not bring Robotina home, it can make robots perform more efficiently around the world, through a dedicated chip that reduces the robot’s response time, which implies The best connection between his body and his “mind”.
It is a robotic computing system, developed by Dr. Sabrina Newman, who holds a degree in Computer Science and Artificial Intelligence from the Massachusetts Institute of Technology and is currently a Fellow at Harvard University. She explains that the construction of robots is increasing, and that their applications are varied, but although most of them show great physical strength, this becomes a challenge for computer systems. Before moving, robots must discover the environment with their sensors, and make decisions in real time, as the electronic and physical components interact. This results in heavy time constraints and computational loads.
Dr. Neumann felt that improving the software was not sufficient to achieve greater efficiency of robots and to explore it with hardware.
MIT collects that the Newman system creates a custom hardware design. The user enters the robot’s parameters – such as the design of its limbs and determines how its joints move – and the system translates those physical properties into mathematical arrays. These values have many zero values that correspond to the impossible motions, due to the particular anatomy of the robot, so the system creates a specialized hardware architecture to run calculations only on non-zero values. The result is a chip designed to maximize the efficiency of each individual robot.
Next month, Newman will present the paper at the International Conference on Architectural Support for Programming Languages and Operating Systems.
Yes, we are still a long way from having robots that are friends, closest, and guides in anyone’s everyday life. But what Newman could have accomplished was to move faster toward those goals, because it would improve the efficiency of existing robots, whose use would be a great relief for humans in risky tasks, such as the attention of patients with infectious diseases, or the load of heavy objects, among other things. .
Founder and Chairman of the Standards Board