Links

Code, competitions and the annual summer and winter camps conducted by our group are hosted on GitHub.

Link to	Description
Multipurpose Household Bot [Multi-Purpose-HouseHold-Bot]	The fundamental concept behind this bot is to build a multifunctional home automation system with a tonne of features that make it affordable for the ordinary household. The majority of household bots available now only perform specific tasks, such as cleaning or playing, but we can add more functionalities without considerably increasing hardware (which won’t increase the cost). The bot will assist with all daily tasks, including child care, cleaning, security, and many other things. Its usability will rise and the price will drop thanks to numerous simple and creative innovations like DIY guides and detachable parts. Also, we’ll make it programmable and customizable by users so they can play around with it, learn from it, and add new features.
3D Occupancy Maps With Quadrotors [3D-Occupancy-Maps-With-Quadrotors]	Simulated Hector Quadrotor In Gazebo and Mapped a Unknown environment. You can map any world using this project. Just replace the new_world in 3D-Occupancy-Maps-With-Quadrotors/drone_test/worlds and rename it new_world. You can create the occupancy map of your gazebo world.
Enhance UnderWater Imagery [Enhancing-Underwater-Imagery-using-Pix2Pix-cGANs]	Autonomous underwater vehicles (AUVs) rely on a variety of sensors for decision-making, with vision-based sensors being one of the most appealing. Nevertheless, the visual data must be improved because red quickly fades and turns the photos blue, and the brightness of the photographs is also impacted by minute variations in the object’s altitude above the seafloor. A noisy and distorted visual data is caused by variables including refraction and absorption, suspended particles in the water, and colour distortion.
Gyrosphere - The Spherical Bot [RoboReG-Gyrosphere]	A ball-shaped mobile robot with spherical external shape typically made of a spherical shell serving as the body of the robot and an internal driving unit (IDU) that enables the robot to move. Gyrosphere typically moves by rolling over surfaces. The rolling motion is commonly performed by changing the robot’s center of mass (i.e., pendulum-driven system).
Gym Environment for Swarm Bots [Swarm-BioNanoRobots]	The challenges in cancer therapies are that they induce side effects by traveling throughout the patient’s body via the bloodstream and destroying both healthy and malignant cells. To reduce the risk of side effects, the therapy should solely target malignancies. Swarm robotics offers the potential to advance personalized precision medicine and cancer therapy.
Pipe Traversal Bot [Pipe-Traversal-Bot]	The challenges involved in cleaning and maintaining Pipes System in various industries that actually require high human workload, health risk and much time consuming. Hence there is much need for such autonomous robot, which can be deployed in various piplines, to ensure robustness and travel to the pipe where it is difficult for a human to reach. Hence, our work is aimed to train the robot to look for the defect and sanitising pipelines which saves money, time and labor effort.
Chase and Run [gym-adversarial-cars]	Adversarial Chase and Run Cars Gym is a gym environment, to test and develop algorithms related to Multi Agent Systems, especially those related to Multi-Agent Reinforcement Learning. This was done under the Robotics Research Group (RoboReG) to try and learn emergent behavior between agents competing against each other via Reinforcement learning and how it could generate new control strategies.
Autonomous Parking [gym-carpark]	The carpark is an environment created using the OpenAI gym and Pybullet, which, as the name suggests, is a car parking arena, where the goal is to autonomously park a car in its parking slot while avoiding collisions. We have done this under the aegis of the RoboReG to see whether a Reinforcement Learning Agent can work in a setting which is not exactly very simple for humans either.
Simultaneous Multitasking Agent [gym-Baxter_SMA]	This environment is made by our team from RoboReG of Robotics Club, IIT(BHU) to tackle an open challenge in the field of robotics, which is Simultaneous work. Generally, Humans have a good efficiency in managing different things simultaneously using both hands. Hence, we want to explore how the Reinforcement Learning Algorithms are trying to achieve this task comparable to humans or even better. For this purpose Baxter humanoid robot is used as the Agent. It is a gym compatible environment made using simulator PyBullet.
Modular Robotics With Swarm Intelligence [gym-iOTA]	This is an Open-AI gym environment developed with a modular bot platform named ‘iOTA’. The motive of this gym is to allow us to test out and develop Algorithms for such a MultiAgent System. This is further used to learn heirarchial planning of such a MultiAgent systems to develop a generalized swarm behaviour in the robots (i.e., Colabortively working towards achieving a objective).
Mask-On [Maskon]	The CORONA virus has taught us many things, right from sanitation of the body to surroundings. So this helps us to find out and keep the area virus free. This project includes using robots for human surveillance in shopping malls and grocery stores.
Waste Autonomous Robotics [Autonomous-Waste-Segregation-Robotics]	Providing Solutions for an Autonomous Waste Management System. Although still a major segment of our population relies on manual labour at dumpyards it cannot be denied that their lives are at a risk and they become prone to health hazards. Many autonomous solutions with implementation of robotics and artificial intelligence have already emerged in industries.

This website is based on code from the SAMPA group at the University of Washington.