Browsing by Author "Upadhyay, Saurabh"
Now showing 1 - 2 of 2
Results Per Page
Sort Options
Item Open Access Autonomous robotic arm manipulation for planetary missions using causal machine learning(European Space Agency (ESA), 2023-10-20) McDonnell, Cian; Arana-Catania, Miguel; Upadhyay, SaurabhAutonomous robotic arm manipulators have the potential to make planetary exploration and in-situ resource utilization missions more time efficient and productive, as the manipulator can handle the objects itself and perform goal-specific actions. We train a manipulator to autonomously study objects of which it has no prior knowledge, such as planetary rocks. This is achieved using causal machine learning in a simulated planetary environment. Here, the manipulator interacts with objects, and classifies them based on differing causal factors. These are parameters, such as mass or friction coefficient, that causally determine the outcomes of its interactions. Through reinforcement learning, the manipulator learns to interact in ways that reveal the underlying causal factors. We show that this method works even without any prior knowledge of the objects, or any previously collected training data. We carry out the training in planetary exploration conditions, with realistic manipulator models.Item Open Access A ROS-based simulation and control framework for in-orbit multi-arm robot assembly operations(European Space Agency (ESA), 2023-10-20) Bhadani, Saksham; Dillikar, Sairaj R.; Pradhan, Omkar N.; Cotrina de los Mozos, Irene; Felicetti, Leonard; Upadhyay, Saurabh; Tang, GilbertThis paper develops a simulation and control framework for a multi-arm robot performing in-orbit assembly. The framework considers the robot locomotion on the assembled structure, the assembly planning, and multi-arm control. An inchworm motion is mimicked using a sequential docking approach to achieve locomotion. An RRT* based approach is implemented to complete the sequential assembly as well as the locomotion of MARIO across the structure. A semi-centralised controller model is used to control the robotic arms for these operations. The architecture uses MoveIt! libraries, Gazebo simulator and Python to simulate the desired locomotion and assembly tasks. The simulation results validate the viability of the developed framework.