Jul 2026· 2026 IEEE/ASME International Conference on Advanced Intelligent Mechatronics (AIM)· pp. 1-8· 0 citations· 30 references
TL;DR
The proposed method is based on a real-time model-based control algorithm used to position a given point belonging to the object, which is grasped by a human and a robot at its endpoints and able to reach the desired position accurately.
Abstract
Developing a human-robot collaborative workplace is the solution to perform faster and more efficient tasks by merging human cognition, awareness, and consciousness with the robot’s power generation, capacity, and precision. In this paper, we address the problem of manipulating linear deformable objects such as cables, ropes, or textiles in a collaborative setup.The proposed method is based on a real-time model-based control algorithm used to position a given point belonging to the object, which is grasped by a human and a robot at its endpoints. The basis of this method lies in (i) the theory of catenaries for modeling the object’s deformation in real-time (ii) the formulation of an interaction matrix representing the robot controller gradient to reach the target position. The experimental results show that the proposed method is reactive to human motion during manipulation and able to reach the desired position accurately.
This work added YOLO-based object and hand detection, stereo vision-based localization using the robot's built-in low-resolution fisheye cameras, and task-specific corrections for grasp execution to form a novel calibration-based grasping pipeline that does not require RGB-D cameras, motion capture, or external tracking systems.
A deep examination of the vision-based object manipulation through collaborative robotics with respect to perception pipelines, object detection and recognition, pose estimation, grasp planning, and real time control integration is given.
Rahul Mehta· International Journal of Int...· 0 citations
Driven by the demands of intelligent manufacturing, medical rehabilitation, and human–robot collaboration, anthropomorphic dexterous hands have become a key direction for complex manipulation. This paper reviews their evolution from structural biomimetics to intelligent control, emphasizing structural optimization, task generalization, and perception–control coupling. First, it summarizes structural design principles, including structural parameterization, underactuated and compliant mechanisms, functional materials, and multimaterial fabrication. Second, it reviews perception technologies, including angle and strain sensing, tactile and pressure detection, and compact vision–tactile fusion modules. Third, it compares classical model-based control, data-driven learning control, and hybrid model–data control, with attention to sample efficiency, robustness, and transferability. Finally, it discusses deployment-oriented strategies, including modular design, standardized interfaces, integrated packaging, and unified communication frameworks for hardware–software coordination. In summary, this paper establishes a system-level technical framework for dexterous hand development and deployment, organized around 4 tightly coupled dimensions: structural embodiment, multimodal perception, control, and deployment-oriented integration. By clarifying the interactions and constraints across these dimensions, the paper provides a system-level basis for the design, evaluation, and deployment of dexterous hands in complex manipulation and human–robot collaboration.
Qianqian Wang, Qijun Yang, Xuanyu An et al.· Research· 0 citations
An adaptive scheme that estimates the kinematic relationship between a robot's joints and the task features it senses online, using only joint-angle sensing and a wrist-mounted force/torque sensor, with no exteroceptive measurement of the tool tip is developed.
Dual-arm robots often encounter difficulties when handling easily deformable or structurally complex objects using traditional grasping-based manipulation. In addition, grasping and releasing operations introduce significant time overhead. To address these limitations, this paper proposes a vision-based predictive control framework for dual-arm nonprehensile transportation. The proposed method employs a hybrid end effector design that integrates an elastic tether with a tray, enabling flexible and stable transportation without direct grasping. A predictive control strategy is adopted to optimize dual-arm motion trajectories on the move under kinematic and safety constraints. To further enhance coordination accuracy, a direct visual servoing scheme is incorporated to dynamically regulate the arm velocities, minimizing relative motion between the end effectors and the object. This effectively suppresses oscillations induced by the elastic tether. Both simulation and experimental results demonstrate that the proposed approach ensures convergence to desired states and achieves continuous, stable, and safe object transportation, even in the presence of disturbances.
Chang Liu, Yuan Yang, Panfeng Huang et al.· 2026 IEEE International Conf...· 0 citations
This paper presents a human–robot interaction (HRI) scheme by using an adaptive admittance control, which helps stroke patients perform rehabilitation training tasks and optimizes their performance. Considering the impact of human factors, the control structure is designed to have two control loops. In the inner loop design, a model‐free adaptive control (MFAC) method is proposed to handle the unmodeled dynamics and unknown disturbances for the desired trajectory tracking, and the convergence and boundedness of this method are strictly proved by using the compression mapping principle. Then, a task‐specific outer loop is developed to find the optimal parameters of the admittance model and transformed into an LQR problem, and a learning algorithm is utilized to solve the given problem without requiring knowledge of the human arm model. Considering the safety of HRI, the constraint of the end‐effector orientation is designed. Simulation studies indicate that the proposed strategy effectively enables stroke patients to execute active training tasks on the robotic exoskeleton.
Unknown authors· International Journal of Rob...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
