4 research outputs found

    A New 2-DOF Dual-Double Tendon-Driven Two-Finger Gripper Design, Analysis, and Performance Evaluation

    No full text
    Robotic dexterity is critical for applications in prosthetics, assistive devices, and industrial automation. Conventional single-DOF grippers lack sufficient joint coordination for precise force distribution, motivating the development of a two-degree-of-freedom (2-DOF) tendon-driven design capable of more human-like articulation. A 2-DOF architecture enables coordinated multi-joint motion and controlled distribution of contact forces, extending functionality beyond the limitations of single-DOF mechanisms and addressing the broader challenge of dexterous manipulation in robotics.This thesis presents the design, implementation, and evaluation of a dual double-tendon, two-finger gripper. A predictive force actuation framework is introduced, combining calibrated load measurements with waveform-derived force estimation to characterise tendon transmission behaviour. A layered control architecture links embedded firmware for real-time acquisition, Python middleware for synchronised logging, and a MATLAB toolchain for kinematic modelling, motion tracking, and fingertip position computation. The analytical mechanism model maps MCP and PIP joint inputs to fingertip coordinates, supporting theoretical validation and trajectory prediction.Mechanically, the gripper design incorporates press fit bearings, refined tendon routing, an improved four-bar linkage, and a re-engineered servo spool geometry to reduce frictional losses, backlash, and hysteresis. Free-body diagram analyses and closed-form kinematics yield fingertip force predictions that show strong alignment with experiments using calibrated loads. Position estimation is enhanced through MATLAB-based edge detection integrated with Kalman-filtered IMU data. Comparative benchmarking demonstrates improved fingertip force output relative to Unde et al. (2023) under matched test conditions.The research contributes a compact and efficient tendon-driven platform supported by systematic analytical, experimental, and comparative evaluation. Beyond establishing a validated design framework, this work highlights pathways for future development, including refined nonlinear models, adaptive closed-loop control, and perception-informed grasp planning through vision and machine learning integration.</p

    CF Patient Cardiac Dynamics (MoCo MSLR: left, Extreme-MRI: right) (supplemental video 7)

    No full text
     CF Reconstructions at high temporal resolution (targeted temporal resolution: 83 ms). Realistic cardiac dynamics can be seen in MoCo-MSLR (left), however there is significant high frequency oscillation present throughout the CINE. This may be because the deformation fields at this extreme level of acceleration are learning to model noise. This high frequency oscillation disrupts the smoothness of the respiratory dynamics which are present but discontinuous. Effectively no motion can be seen in the Extreme MRI reconstruction (right) although some subtle movement of the diaphragm is observed.</p

    Hardness and Chemical Potential Profiles for Some Open Shell HAB → HBA Type Reactions. Ab Initio and Density Functional Study

    No full text
    The electronic structure, hardness (η), and chemical potential (μ) for the 1A‘ and 3A‘‘ states of HNO−HON and the 2A‘‘ state of HSO−HOS have been calculated using HF/6-311++G** and B3LYP/6-311++G** methods. The η and μ profiles of the 1A‘ state of HNO−HON and those of HSO−HOS are obtained in agreement with the salient features of the maximum hardness principle (MHP). However, a quite erratic η profile is predicted for the 3A‘‘ state of HNO−HON. This can be attributed to the nature of the variation in the energy difference of the two states along the reaction path. The relative energies, ionization potentials (I), and electron affinities (A) are calculated at the stationary points of the B3LYP surface using B3LYP and MPn (Full) methods. Most of these values are obtained in very good agreement with the available experimental data. The η values based on these I and A identify the most stable species correctly but do not follow the expected trend with regard to the relative stability of the transition state. The reason for this anomaly is discussed
    corecore