TO: PROFESSOR ELLIS
FROM: ALAIN PALMER
SUBJECT: Enhancing of robots’ abilities using the Cloud, Neuromorphic Computing, and Big Data Analytics.
In this paper “Design of Cloud Based Robots using Big Data Analytics and Neuromorphic Computing” by Satyanarayana, Kusyk, Chen et al discusses some important things about enhancing robots’ abilities. It explained that it is crucial to know the functions and capabilities of the human brain in order to develop robots that will have the ability to link with its environment. Three main functions that was introduced to assist in the enhancement of robots’ abilities are Neuromorphic Computing, Cloud Robots, and Big data Analytics.
Neuromorphic Computing will play a major roll in this project of enhancing robots’ abilities. Neuromorphic Computing is a very large-scale integrating system which included tens to hundreds of transistors per microchip, this was created to imitate the neuro-biological patterns in the nervous system. There has been an increasing demand in neuromorphic computing for its cognitive and brain like computing. With cognitive computing, optimum memory usage, high-speed performances, and low power consumption gives neuromorphic chips its powerful advantages and is also projected to lead the line in rise of robots in the robot industry. There is also a Zeroth processor chip that is created to mimic brain behavior to help robot learn through training and feedbacks instead of hard coding to behave human like.
Second is Cloud Robots. This is the idea to utilize cloud computing. Cloud-Based Robots would be driven by a brain centered cloud for robots. This would work in the way where all robots attached to the cloud will share information that each other can retrieved as soon as it is uploaded to the cloud. A proposed system called RoboEarth is just like humans sharing humans sharing information online, instead it is for robots. Robo earth has an online database for its cloud engine and it is called Rapyuta, which is to perform heavy-weight processing in the cloud. RoboEarth and Rapyuta main job is to process information from various sources and put them in a format that the robots can execute.
Third is the Big Data Analytic. Big data Analytic was first used to indicate large data for visualization. Big Data can enhance and evolve the robot’s Sense, Plan, and Act (SPA) cycle. With that being said, it means that the with Big Data, robots will be more advance in their environment and will be able to predict their next move in any complex situation encountered without being commanded to do so. Quoted (i) “descriptive analytics to understand why something happens, (ii) predictive analytics to predict what could happen, and (iii) prescriptive analytics to suggest what a robot should do in a complex environment (Satyanarayana, Kusyk, Chen et al, 2018 p3). This is what I explained in the previous sentence.
This summary is to highlight the attempt to cover cloud computing, big data analytics, and neuromorphic computing to enhance the control architecture of robotics. In short, to use a very advanced cloud systems to operate robots and decrease the use of the command center.
H. Markram, “Understanding the brain: Organizational and scientific
challenges,” From Physics to Daily Life, pp. 179–190, 2014.
S. Furber, “Large-scale neuromorphic computing systems,” Journal of
Neural Engineering, vol. 13, no. 5, pp. 1–14, 2016.
D. Monroe, “Neuromorphic computing gets ready for the (really) big
time,” Comm. of the ACM Mag., vol. 57, no. 6, pp. 13–15, 2014.
W. Zhao and et al., “Nanotube devices-based crossbar architecture:
toward neuromorphic computing,” Nanotechnology, vol. 21, no. 17, pp.