International Journal of Engineering and Advanced Technology (IJEAT) ISSN: 2249 – 8958, Volume-9 Issue-3, February, 2020 923 Retrieval Number: C5342029320/2020©BEIESP DOI: 10.35940/ijeat.C5342.029320 Published By: Blue Eyes Intelligence Engineering & Sciences Publication Abstract: Robots have been playing a very important role in our day-to-day lives and will be a necessity in the coming future. Whenever we hear automation, the first thing that strikes our mind is a robot performing the given task. But if a robot fails to do the task, it could cost an individual or corporate a huge financial loss. In this study, we have learned the working of various robots and drawbacks that hold them back. For this work, we did make a study of drives used in the robot and after that applied the machine learning algorithms to predict the classification of whether the robot will function properly or not, based on the data of drive(s). Keywords: Machine Learning, Data Science, Robotics I. INTRODUCTION Robots now-a-days have witnessed a large number of users since the process of automation hasbegun. Today, we need Robots in every field, right from agriculture to performing various tasks in corporates. ‘Robolution’ is the term that is used for this era of Robots. But the question arises do we know will the robot perform the given task accurately or even if it does it accurately, does it do it the right way. To overcome such difficulties and ease the human decision making for assigning task to robot, we need algorithms to predict the next result of robot[1], using its dynamic dataset which is integrated with the cloud[2]. The program will be connected to cloud system and will provide output with a prediction graph. The objective of this workis predictive maintenance of industrial robots and the possibility of building a condition monitoring system based on the data analysis of the robot’s locomotion which brings the work of individuals at one place.The aim of this study is also to detect the robot’s accuracy for performing a task correctly.This can be checked by observing a robot for a certain period and then using that dataset for prediction[3]. Revised Manuscript Received on February 04, 2020. Sajal Suhane, Department of Computer Engineering, Dr. D.Y. Patil Institute of Technology, Pimpri, Pune, India. Email: ssuhane31@gmail.com Dr. Pramod D. Patil, Department of Computer Engineering, Dr. D.Y. Patil Institute of Technology, Pimpri, Pune, India. Email: pdpatiljune@gmail.com Ravi Mishra, Department of Computer Engineering, Dr. D.Y. Patil Institute of Technology, Pimpri, Pune, India. Email: mishra.ravi@mail.com Simran Koul, Department of Computer Engineering, Dr. D.Y. Patil Institute of Technology, Pimpri, Pune, India. Email: simran.koul@mail.com Ridima Shukla, Department of Computer Engineering, Dr. D.Y. Patil Institute of Technology, Pimpri, Pune, India. Email: ridimashuklawork@gmail.com Dr. Jyoti Rao, Department of Computer Engineering, Dr. D.Y. Patil Institute of Technology, Pimpri, Pune, India. Email: jyotiasawale@gmail.com Present work majorly focuses on Industrial Robots because their robots perform critical tasks and its failure could be a hefty cost for the industry. Analyzing robot behavior from prior data is the fundamental basis of a data science approach like machine learning. II. LITERATURE SURVEY ”Data Analytics for Predictive Maintenance of Industrial Robots”(Borgi, Tawfik, et al.)presents the predictive maintenance of industrial machines based on the data analysis of robot’s power measurements. A predictive modelling approach is proposed in the paper, to detect robot manipulator accuracy errors based on robot’s current data analysis for predictive maintenance purposes. Also, an experimental procedure is carried out to oversee the correlation between the robot accuracy error and a set of extracted features from current time-series, and to evaluate the proposed predictive modelling. ”Big Data Analytics for Real Time Systems” (Dutta, K., & Jayapal, M.) discusses the use of Big Data and its impact in our day to day lives. It focuses on the fact that Data, if processed at the right time could provide us with some eye- opening insights. An overview of the Big Data Analytics for Real Time Systems and focus on its challenges (3V’s of data), research trends and accuracy of results has been studied in the paper. ”An Empirical Comparison of Supervised Learning Algorithms” (Caruana, Rich, and Alexandru Niculescu- Mizil)comes up with an approach to find the best suited algorithms for Big Data problems using supervised learning methods: SVMs, neural nets, logistic regression, naive bayes, memory-based learning, random forests, decision trees, bagged trees, boosted trees, and boosted stumps. It also examines the effectof calibrating the models via Platt Scaling and Isotonic Regression has on their performance. ”KnowRob—knowledge processing for autonomous personal robots” (Tenorth, Moritz, and Michael Beetz) portrays how data could be processed in the robots in order to get the most out of robot. Knowledge extraction from the given data without hampering the working of robot is one of the major methods discussed in the paper. A Knowledge processing framework and its implementation has been explained which works in almost all the robots (as concluded in the paper). Since, most of the work is done in focused areas such as Robotics, Real time data processing, comparison of algorithms etc. This study tends to bring forward the work done in an interdisciplinary manner for ease of human. Therefore, the work proposes an addition to the existing technology for better robot efficiency and working. Robolution: Real Time Predictive Analytics for Industrial Robots Sajal Suhane, Pramod D. Patil, Ravi Mishra, Simran Koul, Ridima Shukla, Jyoti Rao