Browsing by Author "Rout, Bijay Kumar"

Now showing 1 - 20 of 53

AcYut TeenSize Team Description Paper 2017
(Acyut., 2017) Rout, Bijay Kumar
This paper summarizes the developments in humanoid robotics by Team AcYut of Birla Institute of Technology and Science, Pilani. This paper is submitted as a prerequisite for participation in TeenSize Hu- manoid Soccer League, RoboCup 2017 to be held in Nagoya, Japan. All robots have been developed exclusively by the team. This paper describes the hardware design of our humanoids and the underlying software in its autonomous nature.
Adaptive soft sensor design using a regression neural network and bias update strategy for non-linear industrial processes
(IOP, 2023-05) Rout, Bijay Kumar; Mohanta, Hare Krishna; Pani, Ajaya Kumar
Soft sensing of quality parameters in process industries has been an active area of research for the past two decades. To improve the performance of soft sensors in the scenario of time varying process states, adaptation capability is incorporated into the soft sensor model. In this work, recursive (R), sliding window (SW) and just-in-time learning (JITL) frameworks are used for adaptive soft sensor design. A rarely explored modeling technique in the adaptation framework, the generalized regression neural network (GRNN) is used as a local modeling strategy. A bias update procedure is applied during the model adaptation activity to improve the prediction accuracy. Further, the performances of the developed models are tested against input–output data dimension mismatch along with various concept drift phenomena by considering a different number of labeled samples for inputs and outputs. The proposed adaptation strategy is applied on two benchmark industrial processes. Simulation results show that the GRNN local modeling approach combined with the bias update strategy gives higher prediction accuracy than other adaptive soft sensors proposed in the literature. Moreover, GRNN local modeling strategy using SW adaptation mechanism has the least computation time among the three adaptation methods due to the use of a low number of samples for model development.
Advancements in Yoga Pose Estimation Using Artificial Intelligence: A Survey
(Bentham Science, 2024) Chamola, Vinay; Rout, Bijay Kumar
Human pose estimation has been a prevalent field of computer vision and sensing study. In recent years, it has made many advances that have helped humanity in the fields of sports, surveillance, healthcare, etc. Yoga is an ancient science intended to improve physical, mental and spiritual wellbeing. It involves many kinds of asanas or postures that a practitioner can perform. Thus, the benefits of pose estimation can also be used for Yoga to help users assume Yoga postures with better accuracy. The Yoga practitioner can detect their own current posture in real-time, and the pose estimation method can provide them with corrective feedback if they commit mistakes. Yoga pose estimation can also help with remote Yoga instruction by the expert teacher, which can be a boon during a pandemic. This paper reviews various Machine Learning, Artificial Intelligence-enabled techniques available for real-time pose estimation and research pursued recently. We classify them based on the input they use for estimating the individual's pose. We also discuss multiple Yoga posture estimation systems in detail. We discuss the most commonly used keypoint estimation techniques in the existing literature. In addition to this, we discuss the real-time performance of the presented works. The paper further discusses the datasets and evaluation metrics available for pose estimation.
Analysis of manufacturing supply chain agility performance using Taguchi loss functions and design of experiment
(Emerald, 2018-11) Rout, Bijay Kumar; Sharma, Satyendra Kumar; Routroy, Srikanta
The purpose of this paper is to evaluate the agility performance level of manufacturing supply chains using Taguchi loss functions (TLFs) and design of experiment (DoE).
Applying swing-leg retraction on an actuated spring loaded inverted pendulum model
(IEEE, 2018) Rout, Bijay Kumar
The planar spring-mass model is frequently used to describe bouncing gaits (running, hopping, trotting, galloping) in animal and human locomotion and robotics. Many approximate solutions have been suggested by researchers to predict center of mass trajectory during stance, some of them even proposed models including damping losses. Over the years, many improvements to the model have been suggested with various actuators to take care of non-linearities of the system. But the control laws for those models often require computation after each step. We have used one such model and implemented swing-leg retraction to it, which is an early control method to improve stability of the model. Though a deadbeat control strategy is better, swing leg retraction reduces computation at each step and also the need for precise actuators. We have compared and presented results for the model with and without swing-leg retraction, with control parameters fixed at beginning.
An approach for tracking of mobile robot with vision sensor
(IEEE, 2017) Rout, Bijay Kumar
In mobile robots tracking is an important requirement for real time path planning and obstacle avoidance. Current work proposes an innovative approach which combines feature based object detection, KLT Algorithm based tracking method and Kalman filter based de-noising technique for mobile robot working in real time environment. In the detection phase, the mobile robot is detected using Viola Jones algorithm which extracts detectable feature. Then the position of mobile robot is computed with homography constraints and a region of interest window was set up to accommodate the mobile robot. In the tracking phase, the region of interest window was dealt with using KLT algorithm. The proposed method is of special practical importance in case of specified path tracking as the size of mobile robot image is usually small relative to the captured image of the environment. Thus the analysis of captured image of environment become unnecessary for tracking and thereby reduce computational load. Experiments show that the proposed approach accurately detect and track the mobile robot with error percentage ranging from 0.5% to 8% in different parts of the specified path.
Complete Scene Parsing for Autonomous Navigation in Unstructured Environments
(IEEE, 2020) Rout, Bijay Kumar
Recent developments in Machine Learning and Computer Vision have enabled progress in autonomous navigation. However, most of the existing research focuses on European driving situations, with not much progress made in the Indian context. Our paper aims to achieve complete spatial understanding for the Indian context. Our focus is directed towards Semantic Segmentation and Instance Segmentation. We aim to develop an autonomous navigation pipeline by a combination of both approaches. Using DeepLabv3+ for Semantic Segmentation we achieve an mIOU of 68.58%, and using Hybrid Task Cascade we achieve an Average Recall of 56.5%.
Deep learning based techniques to develop & enhance assistive gear for visually impaired
(IEEE, 2024-12) Rout, Bijay Kumar
This work investigates existing solutions tailored for the visually impaired, focusing on economically viable options for non-first-world communities. The exploration involves developing a real-time obstacle-tracking model using the YOLO (You Only Look Once) algorithm and Text-to-Speech synthesis to provide auditory cues. This effort yields improvements in assistive technology, though it still faces limitations in algorithmic precision and user feedback integration. The research paves the way for refining this technology and envisions its seamless integration into the daily lives of the visually impaired. The findings enhance the performance of assistive technologies, especially for distances less than 1.5 meters. The results show an inaccuracy of less than 10%, translating to a margin of 10−15 cm for objects located one meter away. This work thus provides increased independence and confidence for individuals with visual impairments in navigating and interacting with their surroundings.
Design and Trajectory Optimization of Delta Robot
(Springer, 2021-07) Rout, Bijay Kumar
Delta robots are parallel robots that allow the movement of an end-effector platform parallel to a base. These robots are extensively used for pick-and-place applications in the packaging industry and high precision assembly operations in electronics and 3-D printers as these have capability for high-speed operations. The robot has 4-DOF with three translational joints and one rotational joint. The paper attempts to determine an optimum path, given the start and end points of the end-effector of a virtual delta robot, and to minimize the energy consumed by the motors to drive the platform. In addition, it synthesizes a configuration that consumes the least energy while traversing a given path
Development of a Shape Aware Path Planning Algorithm for a Mobile Robot
(IEEE, 2020) Rout, Bijay Kumar
A shape aware path planning algorithm is necessary for real time execution of a task by a mobile robot. Current work proposes a shape-aware A* path planning approach to facilitate accurate path finding in a given environment to accommodate the shape of mobile robot for differential wheeled mobile robot. The real-time map allows to assign a favorable cost value for each grid location of the map which later used to develop a shape aware global path planning strategy by using the well-known A* algorithm. For implementation and validation, an overhead camera is used to capture the task space and the obstacles which work in Robot operating software platform. The proposed method was tested in a real-time environment and proved the algorithm is capable of moving in a path that minimizes the distractions to static obstacles.
Effective localization of humanoid with fish-eye lens using field line detection
(IEEE, 2016) Rout, Bijay Kumar
This paper presents a real-time technique for efficient localization of a humanoid robot in a soccer field using a fish-eye lens. We propose a novel method that uses node merging to provide a robust and an accurate estimate of positions of the intersection points in the field lines. The distances to these points are calculated using Inverse Perspective Mapping and fed to the particle filter for the purpose of localization. The proposed algorithms have been implemented and tested on the humanoid robot platform AcYut, being developed at the Centre for Robotics and Intelligent Systems (CRIS), BITS Pilani.
Energy efficient reactionless design of multi-arm space robot for a cooperative handshake maneuver
(Elsevier, 2023-02) Rout, Bijay Kumar
Space robots play a significant role in on-orbit capture, space structure construction, and assembly tasks. Since the robotic arms are attached to a free-floating satellite, the motion of the manipulator in such tasks and the satellite are coupled. Multiple-arm space robots can perform complex cooperative tasks and are superior to single-arm space robots. Current work proposes a reactionless manipulation algorithm for a multi-robotic arm based on the iterative Newton–Euler method for space robots with many task and balance arms. The present work demonstrates two tasks and one balance arm to perform a reactionless handshake maneuver in space. This maneuver is presented in detail for a planar and spatial case. The planar case uses 3 DoF robotic arms, while the spatial case uses 6 DoF robotic arms. In addition, the balance arm has been designed considering the efficient usage of energy satisfying reactionless manipulation concept. The design procedure focuses on minimizing energy used during the motion of the balance arm for a known motion of task arms using a genetic algorithm. Moreover, computational experiments are conducted to validate the use of the genetic algorithm for optimization. The results of proposed reactionless manipulation algorithm have been validated with the results available in the literature for the spatial case that uses a different method. In the future, an energy-efficient balance arm will be designed to handle tumbling objects.
A FEM and Image Processing Based Method for Simulation of 01 Manufacturing Imperfections
(ARME, 2012) Rout, Bijay Kumar
Use of appropriate methods to capture manufacturing imperfection at the conceptual stage is a major challenge for the designer and researchers in industry. Imperfections are observed in almost all type of in macro, micro and nano-machining domain of manufacturing process. These imperfections lead to undesirable performance in application phase. In the present work, a simulation based approach to handle manufacturing imperfection is implemented using image processing operators. This method simulates the image of the component due to manufacturing imperfections. The usage of these image processing operators facilitates a realistic simulation of manufacturing errors, in macro, micro, and nano domain manufacturing. The simulated image is further processed for its structural properties i.e. maximum deflection, reactions, Von Mises stress, and change in amount of material, corresponding to its intended application. In order to generate these results based on modified image of beam, the concept of "Solid Isotropic Material with Penalization"(SIMP) is utilized along with 2-D finite element routine. An example of a simple cantilever beam is selected to illustrate the proposed methodology, and the results are analyzed. The present work discusses a simple and easy method to predict the behavior of designed component prior to its manufacturing.
Geometrical approach to On line Trajectory generation, Obstacle avoidance and Footstep planning for a Humanoid Robot
(IEEE, 2015) Rout, Bijay Kumar
Abstract—This paper presents a unique real time technique for path generation in dynamic environment using simple geometry, suitable for humanoid robots. We show that the proposed algorithm reduces computation to find the optimal collision free path by utilizing the geometrical shapes of the obstacles to navigate from a known initial point to final point. The trajectory is then translated to known parameters of humanoid gait model thus effectively computing the footsteps for the robot. This algorithm has been simulated, implemented and tested on humanoid robot AcYut, developed at Centre for Robotics and Intelligence Systems, BITS-Pilani.
Hyper-parameter Optimization on Viola Jones Algorithm for Gesture Recognition
(Springer, 2020-07) Rout, Bijay Kumar
The problem of features, objects, gestures, and face detection has been tackled using a numerous vision-based algorithms available in literature. Each of these algorithms requires a set of hyper-parameters, which need to be set on the basis of trial and error such that the results provide best performance to a situation. Mostly, researchers use trial and error approach to satisfactory result and solve the above problems. In this work, an approach has been suggested to determine an optimum set of hyper-parameters, which will provide a starting point for anyone using Viola Jones algorithm for hand gesture recognition or similar endeavors. This will reduce the time spent in searching for the best combination of hyper-parameters.
Implementation of cascade classifier technique for gesture control of an ABB industrial manipulator
(IEEE, 2018) Rout, Bijay Kumar
There have been several attempts already made to control an industrial robot using hand gestures. The current work proposes a different methodology for controlling an industrial robot. Instead of having an external arrangement of controllers, the RobotStudio software which is compatible and intended for ABB robots, is maneuvered to achieve gesture control of robot by connecting it with python programming software. Here, a Python programming software with an OpenCV library is used to acquire coordinates of the hand gesture recognized with the help of Cascade Classifier technique. We have implemented Socket Programming between RobotStudio software and Python software for transmitting those coordinates to the robot and thus accomplish a real time interface between them.
Influences of design tools on the original and redesign processes
(Taylor & Francis, 2013-08) Rout, Bijay Kumar
Concept generation plays a vital role in establishing a broader foundation in the design process to create novel products. In the globalized, collaborative, designing scenario, an unambiguous representation of captured ideas to explicate a designer's thoughts is important in the sharing and reuse of concepts. Various design studies have noted the impact of design tools on concept generation. However, the results did not detail the influences of a variety of tools on the representation and reinterpretation of concepts through captured design documents. The goal of this paper was to understand the influence of conceptual design tools: Hi-Tech®Mobile e-Notes TakerTM, Wacom® Tablet, and Rhinoceros® CAD with MS Word/PowerPoint on concept representation and reinterpretation, during the original and redesign phases. Eighteen design experiments, involving six individual student designers' solving three design problems each, were conducted in the original and redesign phases. The analyses of 26 variables from captured documents and video protocols reveal that the design tools had a statistically significant impact on four key variables: the total time taken to solve each problem, the time spent on detailed design activity, the textual representation of structural requirements, and the graphical representation of the structure of detailed concepts. Irrespective of the design tool used, novice designers generated a low number of redesign concepts. This makes us conclude that designers might require training for reinterpretation and extracting necessary information from the concepts originally captured, rather than working with poor understanding, ambiguity, and assumptions about the original designer's intent.
Integrated method for performance analysis of reliability-based topologically optimized components
(Sage, 2019-08) Rout, Bijay Kumar
The available robust and reliable topology optimization methods provide quick and efficient design output in an uncertain environment. However, the whole domain of performance function remains hidden during this design process. In the interest of the designer, it is required to know the overall behavior of performance functions in deterministic as well as uncertain/realistic environment. The current work achieves this by proposing an integrated methodology, which combines the design of experiments approach and reliability-based topology optimization. The proposed method enables the designer to simulate performance functions in a desired design-factors space, including uncertainties, via reliability value. For this analysis, compliance, maximum deflection, mechanical advantage, and von Mises stress values are selected as performance functions. Volume fraction, applied force, and dimensions or aspect ratio are chosen as design/control factors. The uncertainties of these design factors are captured using reliability-based topology optimization. The uncertainties due to noncontrollable factors such as material property, load direction, and magnitude are incorporated using the design of experiments approach. Under these uncertainties, the performance of topologically optimized problem is simulated for different experimental combinations of the design factors. The experimental combinations for uncertainties and design factors are generated using Taguchi's orthogonal array. Simulated results are analyzed using techniques such as analysis of mean and variance, signal-to-noise ratio, and response surface method. These analyses help in identifying statistical significance of factors and uncertainties, performance variations, and equivalence relation of performance vs. factor. The proposed methodology is illustrated by selecting monolithic structures such as, on MBB, cantilever beam, and force inverter mechanism.
An integrative approach for path planning and tracking of a shape-aware mobile robot in structured environment using vision sensor
(Inder Science, 2021-04) Rout, Bijay Kumar
A shape-aware path planning algorithm is necessary for real-time execution of a task by a mobile robot whereas path planning algorithms available in literature consider mobile robot as a point object. Current work proposes a shape-aware A* path planning approach with a heuristic function to accommodate the shape of mobile robot. In this paper, the detection, tracking and control of the robot has been carried out for the mobile robot while performing a task in a structured environment. For the implementation and validation, an overhead camera is used to capture the images of obstacles in the task space and published in ROS platform. The captured images are also processed using OpenCV software for detection and tracking using Kanade-Lucas-Tomasi (KLT) and Kalman filter algorithms for different test scenarios. The proposed approach accurately detects and tracks the shape aware mobile robot with percentage error ranging from 6%-10% in different cases.
An integrative approach for tracking of mobile robot with vision sensor
(Inder Science, 2019-03) Rout, Bijay Kumar
Current work addresses an experimental approach which incorporates feature-based object detection, KLT algorithm-based tracking method and Kalman filter-based de-noising technique in a real-time environment. In the detection phase, the mobile robot is detected using Viola-Jones algorithm which extracts detectable features. Then the position of the mobile robot is computed with homography constraints and a region of interest window is set up to accommodate the mobile robot. In the tracking phase, the region of interest window is dealt with using KLT algorithm. The proposed method is of practical importance when the mobile robot is tracked while moving on a predetermined (specified) path as the size of the image of the mobile robot is small relative to the captured image of the environment. Thus the analysis of captured image of environment becomes unnecessary for tracking and thereby the approach reduces computational load. The proposed approach accurately detects and tracks the mobile robot with error percentage ranging from 0.5% to 10% in different parts of the specified path.