mayankJoshi's Projects
A 3D Object detection pipeline using 3D LiDAR points
Implementation of A star path planning
AR Tag Detection and Superimpose virtual objects on it
ROS beginner tutorials
BlAInder range scanner is a Blender add-on to simulate Lidar and Sonar measurements. The result can be saved as annotated 2D image or 3D point cloud.
A complete computer science study plan to become a software engineer.
Python based implementation of the paper "Controlling Shapes of Ensembles of Robots"
Simple cpp boilerplate with cmake and gtest
YOLOv4 / Scaled-YOLOv4 / YOLO - Neural Networks for Object Detection (Windows and Linux version of Darknet )
Implemented Dijkstra algorithm to plan a path between two points
Custom Face Recognition Application, scans the user-provided database for faces and then matches it with detected faces in the video stream
Attach two Gazebo models with a virtual joint in a generalized grasp hack
computer-vision based hand recognition virtual calculator application
A C++ module for Object Detection and tracking (in this case humans are detected, but 80 classes from COCO dataset can be detected). The coordinates of all the objects detected in the camera frame are transformed into robot reference frame for further processing.
Mobile manipulator which can autonomously navigate in a restaurant environment while serving food to the customers
Config files for my GitHub profile.
Final Destination
Designing LQR and LQG controller for a non-linear dual suspended load system
A ROS C++ pacakge for Swarm of 20 robots(TurtleBot3) to perform a search and rescue operation. Each robot navigates autonomously to a designated waypoint, searches for humans, if a human is detected, plans a path to the nearest fire exit, and then returns back to its home location.
My undergraduate project, the main objective was to design a low-cost robotic manipulator having 2 Degrees of Freedom and a gripping mechanism
Some basic ROS projects: talker-listener, turtlesim-motion, turtlebot-laser
3D scene reconstruction and simultaneously obtain the camera poses with respect to the scene, using Linear Triangulation and PnP. Levenberg Marcqdat optimization was done using Reprojection error cost function to optimize for the depth and pose estimates. Project 3 of the course CMSC733@UMD.
Pytorch version of SfmLearner from Tinghui Zhou et al.
Trained a deep neural network to classify traffic signs using German Traffic Sign Dataset
A simple turtleBot3 walker algorithm