Whole-Body Proprioceptive Morphing: A Modular Soft Gripper for Robust Cross-Scale Grasping

TL;DR

This work tackles the fixed-morphology limitation of conventional soft grippers by introducing proprioceptive morphing, a distributed, modular, self-sensing actuation network that enables whole-body reconfiguration. The gripper combines four morphing palm actuators with four bending grasping fingers, all 3D-printed into a low-cost, scalable platform that provides closed-loop control through embedded proprioception. Experimental validation shows a substantial expansion of the grasping envelope across diverse shapes and scales (up to ~10x), as well as novel capabilities like multi-object grasping and internal-hook grasps. Collectively, the method offers a practical path toward biologically inspired, dexterous manipulation that is easy to fabricate and adapt for real-world tasks.

Abstract

Biological systems, such as the octopus, exhibit masterful cross-scale manipulation by adaptively reconfiguring their entire form, a capability that remains elusive in robotics. Conventional soft grippers, while compliant, are mostly constrained by a fixed global morphology, and prior shape-morphing efforts have been largely confined to localized deformations, failing to replicate this biological dexterity. Inspired by this natural exemplar, we introduce the paradigm of collaborative, whole-body proprioceptive morphing, realized in a modular soft gripper architecture. Our design is a distributed network of modular self-sensing pneumatic actuators that enables the gripper to intelligently reconfigure its entire topology, achieving multiple morphing states that are controllable to form diverse polygonal shapes. By integrating rich proprioceptive feedback from embedded sensors, our system can seamlessly transition from a precise pinch to a large envelope grasp. We experimentally demonstrate that this approach expands the grasping envelope and enhances generalization across diverse object geometries (standard and irregular) and scales (up to 10$\times$), while also unlocking novel manipulation modalities such as multi-object and internal hook grasping. This work presents a low-cost, easy-to-fabricate, and scalable framework that fuses distributed actuation with integrated sensing, offering a new pathway toward achieving biological levels of dexterity in robotic manipulation.

Figures (10)

• Figure 1: Motivation. (a) The octopus uses its flexible tentacles for masterful multi-scale object sensing and manipulation, providing our biological inspiration. (b) A rigid gripper fails to conform to the object, resulting in an unstable grasp. (c) A conventional soft gripper, limited by its fixed structure, cannot handle objects outside its designed size range. (d) Our proposed gripper uses adaptive, whole-body shape morphing to reconfigure its entire structure, enabling it to securely envelop objects of varying scales. (e) Dynamic grasping sequence: the gripper first adapts its global shape during approach and then performs a final enveloping grasp for a secure lift.
• Figure 2: System architecture for whole-body proprioceptive morphing. (a) The modular mechanical design consists of four morphing palm actuators that control the gripper's global shape and four grasping finger modules for object envelopment. These are coupled by rigid connectors to form a reconfigurable structure. (b) The hierarchical control and sensing schematic. A central microcontroller runs parallel control loops for the finger and morphing modules. Proprioceptive feedback is achieved via integrated bend sensors in the fingers and pressure sensors in the morphing actuators. (c) Closed-loop pressure control logic for a single morphing actuator. A high-level command for a desired length ($L_{des}$) is translated by a PID controller into a target pressure ($P_{target}$). A low-level logic then modulates pneumatic valves (inflate, hold, deflate) to precisely regulate the actuator's state.
• Figure 3: Single actuator characterization. The plots show a predictable relationship between applied pressure and the resulting (a) length of the palm actuator and (b) bending angle of the finger actuator. Illustrations show the corresponding shape morphing transitions: the palm extends from 68 mm to 135 mm to reconfigure and morph the gripper's framework shape, while the finger bends inward to grasp objects.
• Figure 4: Shape manifold characterization of the framework formed by the four palm actuators. Interpolated heatmaps illustrating the palm module's ability to achieve various target shapes. For each shape – (a) kite, (b) rectangle, and (c) trapezoid – the heatmaps show the minimum internal angle (color bar) as a function of adjustable link lengths (x and y axes in millimeter scale, as defined in the accompanying diagrams). The 'x' marks indicate the measured points used for interpolation.
• Figure 5: Fabrication process. (a) All parts are fabricated using two materials, TPU and PLA, and printed using a Bambu Lab X1 printer. The entire set can be printed simultaneously on a single 256 mm × 256 mm build plate over approximately 36 hours of printing. After 3D printing, all components are assembled using screw joints, except for the friction pads at the fingertip, which are bonded with silicone epoxy (Silpoxy). (b) Cross-sectional views of the finger actuator and palm actuator are shown. The yellow-shaded regions indicate the sensor housings, and pneumatic pressure is applied to both actuators through the air inlet ports. The holes at the distal ends correspond to M2 screw locations used for fastening.