cs.ROMar 29, 2026

Probe-to-Grasp Manipulation Using Self-Sensing Pneumatic Variable-Stiffness Joints

Authors: Ngoc Duy Tran, Yeman Fan, Feng Dai, Khang Nguyen, Anh Nguyen, Hoang Hiep Ly, Tung D. Ta, Shigeru Chiba

Read PDF
Ask AI
Cite

Copy citation

DOINot available for this paper

Abstract

Grasping deformable objects with varying stiffness remains a significant challenge in robotics. Estimating the local stiffness of a target object is important for determining an optimal grasp pose that enables stable pickup without damaging the object. This paper presents a probe-to-grasp manipulation framework for estimating the relative stiffness of objects using a passive soft-rigid two-finger hybrid gripper equipped with self-sensing pneumatic variable-stiffness joints. Each finger of the gripper consists of two rigid links connected by a soft pneumatic ring placed at the joint, enabling both compliant interaction and controllable joint stiffness via internal pressurization. By measuring the pressure inside the pneumatic ring, we can estimate the interaction force during contact. Building on this, we propose a practical probing strategy to infer relative object stiffness by correlating the estimated normal force with known gripper closing displacement. We validate the self-sensing model through stiffness characterization experiments across bending angles and pressure ranges, and demonstrate stiffness-aware probing-and-grasping in real-life applications: selecting grasp locations on fruits with spatially varying stiffness. The proposed system offers a minimal, low-cost sensing approach for stiffness-aware soft manipulation while retaining probing and grasping capability.

Figures & tables

Fig. 1: Overview of the hybrid gripper mounted on a robotic arm during grasping and probing tasks. (a) The gripper integrates rigid links with a pneumatic soft-ring joint, enabling compliant interaction and variable stiffness control. (b) The hybrid gripper performs a probing process on elongated fruits, such as a banana, to estimate stiffness before grasping. (c) For round fruits such as an orange, the gripper rotates around the object to probe different locations and identify the stiffest region for a stable grasp.
Fig. 2: Hardware Details: Structural components and dimensions of the newly designed finger and soft pneumatic system. (a) The pneumatic system consists of a pressure sensor and a soft ring, both controlled through the same air channel connected to a solenoid valve. (b) Cross-sectional view of the hybrid finger showing the newly designed fabric-attaching thread mechanism integrated into the finger body. (c) The drawing illustrates the cross-sectional dimensions of the soft core, which includes an additional tube designed to simplify the installation of the silicone tube.
Fig. 3: System hardware: Experimental platform comprising an xArm 7 robotic arm equipped with a hybrid gripper, a pneumatic supply line with a pressure regulator (SMC CORPORATION), and a control board responsible for command generation, signal interfacing, and overall system integration.
Fig. 4: Demonstration of the data-collection procedure used to validate stiffness estimation. A force gauge mounted on the robot arm follows a circular trajectory around the joint as the fingertip bends. For each bending angle, the torque \uptau\uptau generated by the soft ring is computed from the tangential force FtF_{t} and the moment arm length.
ObjectsRelative Stiffness (N/mm)
Cube 1 (20A)50.83
Cube 2 (30A)54.87
Cube 3 (80A)202.39
TABLE I: Relative stiffness of tested objects
Fig. 5: Pressure variation during the probing process under different initial pressures. Each subplot shows the relationship between the change in pressure and the gripper closing distance for objects with different stiffness levels (Cube 20A, Cube 30A, and Cube 80A). As the closing distance increases, the pressure change becomes more significant for stiffer objects, demonstrating that pressure variation can be used to distinguish relative object stiffness. In addition, the gripper demonstrates higher sensitivity in differentiating stiffness at an initial pressure of 60 kPa60\text{\,}\mathrm{kPa} and a closing distance of approximately 30 mm30\text{\,}\mathrm{mm} .
Fig. 6: The torque was measured in soft-joint stiffness experiments, with pressure continuously controlled by a regulator from 0 to 150 kPa and bending angles ranging from 0 \SIUnitSymbolDegree0\text{\,}\mathrm{\SIUnitSymbolDegree} to 60 \SIUnitSymbolDegree60\text{\,}\mathrm{\SIUnitSymbolDegree} .
Fig. 7: The torque was measured in soft-joint stiffness experiments under constrained air-volume conditions by solenoid valves, with bending angles ranging from 0 \SIUnitSymbolDegree0\text{\,}\mathrm{\SIUnitSymbolDegree} to 60 \SIUnitSymbolDegree60\text{\,}\mathrm{\SIUnitSymbolDegree} and pressures ranging from 0 to 80 kPa.
Fig. 8: (a) Torque–bending angle relationship during Phase 1 (increasing) and Phase 2 (decreasing). (b) Corresponding pressure variation as a function of bending angle for both phases. In Phase 1, data were recorded as the fingertip bent progressively from 0 \SIUnitSymbolDegree0\text{\,}\mathrm{\SIUnitSymbolDegree} to a specified angle. In Phase 2, data were recorded as the fingertip bent to a maximum angle of 80 \SIUnitSymbolDegree80\text{\,}\mathrm{\SIUnitSymbolDegree} and then returned to the specified angle, where measurements were taken.
Fig. 9: The relative stiffness of the banana was measured at different positions along its length. The softest region coincided with the black spot on the peel.
Fig. 10: The relative stiffness of the semi-rooted orange was measured at different positions around its center. Based on these results, the orange should be grasped at the position with the highest stiffness to ensure a stable grip.
More in this topic

Robotic Grasping

Robotic grasping studies how robots perceive, plan, and execute stable, task-appropriate contacts with objects across varied shapes and conditions.

See all 311 papers

Explore similar work

CardsList

  1. A Human-Inspired Thumb-Index Robotic Hand with Strain Gauges Embedded in Soft Joints

    Jun 19, 2026Jonas Papenbrock, Shubhan Patni, Tomaso Lisini Baldi +4Robotic Hand DesignRobotic Grasping

    Department of Information Engineering and Mathematics - University of Siena. · Department of Cybernetics, Faculty of Electrical Engineering, Czech Technical University in Prague

  2. Relaxation-Aware Multimodal Sensing of Soft Gripper Driven by Structure-Perception-Learning

    Aug 27, 2026Yanzhe Wang, Hao Wu, Ziyi Zheng +1Robotic GraspingSoft Robotics

    Grasp Lab, School of Mechanical Engineering, Zhejiang University · Torch Kernel Co., Ltd.

  3. Monocular Vision Based Control Framework for Grasping

    Jul 8, 2026Shail Jadav, Dongheui LeeRobotic GraspingDeformable Object Manipulation

    Autonomous Systems, Technische Universität Wien (TU Wien), Vienna, Austria · Institute of Robotics and Mechatronics (DLR), German Aerospace Center, Wessling, Germany