Embedded Files
Publications
Publications
Suman Raj, Swapnil Padhi, Ruchi Bhoot and Prince Modi, "Towards Collision Avoidance for UAVs to Guide the Visually Impaired”, Late-Breaking Results Abstracts, IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2023
Suman Raj, Harshil Gupta and Yogesh Simmhan, "Deadline-driven Heuristics for Scheduling DNN Inferencing on Edge and Cloud for UAV Fleets", IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing (CCGRID), 2023
Suman Raj and Yogesh Simmhan, “Mobile app platform for Personalized UAV Fleets using Edge and Cloud”, Early Career and Students’ Showcase Poster at IEEE/ACM International Symposium on Cluster, Cloud and Internet Computing (CCGRID), 2023
Shashwat Jaiswal, Suman Raj, Subhajit Sidhanta and Yogesh Simmhan, "A Lightweight, Mobility-Aware, Geospatial & Temporal Data Store for Multi-UAV Systems", Early Career and Students’ Showcase Poster at IEEE/ACM International Symposium on Cluster, Cloud and Internet Computing (CCGRID), 2023
Suman Raj, Swapnil Padhi and Yogesh Simmhan, "Ocularone: Exploring Drones-based Assistive Technologies for the Visually Impaired", Late Breaking Work (LBW), Extended Abstracts of the CHI Conference on Human Factors in Computing Systems (SIGCHI), 2023
Suman Raj, Harshil Gupta and Yogesh Simmhan, “Real-Time Edge Analytics of Video Feeds from a UAV Fleet”, Student Research Symposium Poster, IEEE International Conference on High Performance Computing, Data, and Analytics (HiPC), 2021
Mukund Mitra*, Suman Raj* and Shishir Kolathaya, Inverse Kinematics Analysis of Cassie Robot using Radial Basis Function Networks, International Symposium of Asian Control Association on Intelligent Robotics and Industrial Automation (IRIA), 2021 (Best Paper Award)
Selected Course Projects
Selected Course Projects
CP230_Project_Report (1).pdfPath Planning for Polyhedral Robots using Connectivity Grids, IISc (2022)
Motion planning for polyhedral robots is time consuming. Complexity of traditional planning algorithm increases exponentially for higher degrees of freedom robot. Significant research is done on motion planning for translating polyhedral robot with lesser emphasis on rotation. In this paper we present an algorithm for path planning of polyhedral robot executing translational and rotational motion in complex 3D environment. Forbidden space for the polyhedral robot is formulated by using minkowski sum of robot and the obstacles. We construct a roadmap by performing an adaptive grid volumetric sampling of the workspace. This allows us to capture the connectivity of the free space and perform exact path planning.
Motion planning for polyhedral robots is time consuming. Complexity of traditional planning algorithm increases exponentially for higher degrees of freedom robot. Significant research is done on motion planning for translating polyhedral robot with lesser emphasis on rotation. In this paper we present an algorithm for path planning of polyhedral robot executing translational and rotational motion in complex 3D environment. Forbidden space for the polyhedral robot is formulated by using minkowski sum of robot and the obstacles. We construct a roadmap by performing an adaptive grid volumetric sampling of the workspace. This allows us to capture the connectivity of the free space and perform exact path planning.
3DoF_Robotic_Arm (1).pdfAnalysis of a 3DoF Robotic Arm, IISc (2021)
The manipulability and the dexterity of any robotic manipulators depend upon its degree of the redundancy. Serial robotic arm is very popular in industrial applications because of its simplistic designs. Serial robotic manipulators are also designed for the joint fault tolerance. In this project we analyze the forward and inverse kinematics, forward and inverse dynamics, trajectory generation of a 3 DOF serial robotic arm. The analytical results have been compared with the simulations done on RoboAnalyzer© software. The results were in quite agreement.
The manipulability and the dexterity of any robotic manipulators depend upon its degree of the redundancy. Serial robotic arm is very popular in industrial applications because of its simplistic designs. Serial robotic manipulators are also designed for the joint fault tolerance. In this project we analyze the forward and inverse kinematics, forward and inverse dynamics, trajectory generation of a 3 DOF serial robotic arm. The analytical results have been compared with the simulations done on RoboAnalyzer© software. The results were in quite agreement.
As an extension, we have worked on transforming task space torque to configuration space torque for a 2R planar manipulator and verified it using ode solver.
As an extension, we have worked on transforming task space torque to configuration space torque for a 2R planar manipulator and verified it using ode solver.
FYP_Thesis__Copy_ (1).pdfCollision Avoidance of Multi Agent Systems,
NITT (2019)
We have developed an algorithm for the construction of distant optimal paths and collision free trajectories from the paths for the multi-robots to navigate the environment. The path planning has been carried out with three different algorithms – Dijkstra , A* and D* lite algorithm. A comparison study has been made to find the best suitable algorithm for finding the optimal path to be used in the trajectory generation. The generation of a time parameterized path (trajectory) is carried out using Cubic spline Interpolation. Non-holonomic robots with motion constraints are considered because of their wide application in fields like Kiva systems and Self driving cars. The algorithm has been implemented on four robots and can be extended to N-robots. The testing of the algorithm has been carried out on the simulation environment in MATLAB.
We have developed an algorithm for the construction of distant optimal paths and collision free trajectories from the paths for the multi-robots to navigate the environment. The path planning has been carried out with three different algorithms – Dijkstra , A* and D* lite algorithm. A comparison study has been made to find the best suitable algorithm for finding the optimal path to be used in the trajectory generation. The generation of a time parameterized path (trajectory) is carried out using Cubic spline Interpolation. Non-holonomic robots with motion constraints are considered because of their wide application in fields like Kiva systems and Self driving cars. The algorithm has been implemented on four robots and can be extended to N-robots. The testing of the algorithm has been carried out on the simulation environment in MATLAB.
Page updated
Google Sites
Report abuse