arXiv subjects

Berke Gur

Publications and source records attributed to Berke Gur.

3 recordsLinked to original sources

arXiv2025

A Robot Simulation Environment for Virtual Reality Enhanced Underwater Manipulation and Seabed Intervention Tasks

This paper presents the (MARUN)2 underwater robotic simulator. The simulator architecture enables seamless integration with the ROS-based mission software and web-based user interface of URSULA, a squid inspired biomimetic robot designed for dexterous underwater manipulation and seabed intervention tasks. (MARUN)2 utilizes the Unity game engine for physics-based rigid body dynamic simulation and underwater environment modeling. Utilizing Unity as the simulation environment enables the integration of virtual reality and haptic feedback capabilities for a more immersive and realistic experience for improved operator dexterity and experience. The utility of the simulator and improved dexterity provided by the VR module is validated through user experiments.

cs.RO↗

arXiv2025

Project URSULA: Design of a Robotic Squid for Underwater Manipulation

With this paper, the design of a biomimetic robotic squid (dubbed URSULA) developed for dexterous underwater manipulation is presented. The robot serves as a test bed for several novel underwater technologies such as soft manipulators, propeller-less propulsion, model mediated tele-operation with video and haptic feedback, sonar-based underwater mapping, localization, and navigation, and high bandwidth visible light communications. Following the finalization of the detailed design, a prototype is manufactured and is currently undergoing pool tests.

cs.RO↗

arXiv2022

Kinematics Transformer: Solving The Inverse Modeling Problem of Soft Robots using Transformers

Soft robotic manipulators provide numerous advantages over conventional rigid manipulators in fragile environments such as the marine environment. However, developing analytic inverse models necessary for shape, motion, and force control of such robots remains a challenging problem. As an alternative to analytic models, numerical models can be learned using powerful machine learned methods. In this paper, the Kinematics Transformer is proposed for developing accurate and precise inverse kinematic models of soft robotic limbs. The proposed method re-casts the inverse kinematics problem as a sequential prediction problem and is based on the transformer architecture. Numerical simulations reveal that the proposed method can effectively be used in controlling a soft limb. Benchmark studies also reveal that the proposed method has better accuracy and precision compared to the baseline feed-forward neural network

cs.RO↗