Surface-based Manipulation Using Tunable Compliant Porous-Elastic Soft Sensing
Authors: Gayatri Indukumar, Muhammad Awais, Diana Cafiso, Matteo Lo Preti, Lucia Beccai
Organizations: Soft BioRobotics Perception Group at Istituto Italiano di Tecnologia, 16163, Genova, Italy · University of Genoa, 16126, Genova, Italy · Biorobotics Institute, Scuola Superiore Sant’Anna, Pontedera, 56025, Italy · Soft Robotics Lab, National University of Singapore, 117575, Singapore
There is a growing need for soft robotic platforms that perform gentle, precise handling of a wide variety of objects. Existing surface-based manipulation systems, however, lack the compliance and tactile feedback needed for delicate handling. This work introduces the COmpliant Porous-Elastic Soft Sensing (COPESS) integrated with inductive sensors for adaptive object manipulation and localised sensing. The design features a tunable lattice layer that simultaneously modulates mechanical compliance and sensing performance. By adjusting lattice geometry, both stiffness and sensor response can be tailored to handle objects with varying mechanical properties. Experiments demonstrate that by easily adjusting one parameter, the lattice density, from 7 % to 20 %, it is possible to significantly alter the sensitivity and operational force range (about -23x and 9x, respectively). This approach establishes a blueprint for creating adaptive, sensorized surfaces where mechanical and sensory properties are co-optimized, enabling passive, yet programmable, delicate manipulation.
Figures & tables
Figure 1: Overview of the proposed COPESS-based surface manipulation system. (a) Photograph of the COPESS-based surface manipulation setup, consisting of a sensing unit with a 4×4 coil array, a lattice-based modulating layer, and a target layer. (b) Schematic illustration of the COPESS-based manipulation mechanism. The sensorized tile is covered with a 3D-printed lattice structure, which acts as a modulating layer. As the object transitions from a high-stiffness lattice to a low-stiffness lattice region, its velocity decreases while the sensor response increases.
Figure 2: Mechanical and electrical characterization informing the sensor’s design. (a) Force vs. displacement curves of three different lattice structures, demonstrating that the Gyroid (blue) offers the best combination of initial stiffness and strain range before densification. (b) Comparing a gyroid lattice structure printed using Elastic 50AV2 resin against the characteristic curve of a typical hyperelastic material. (c) Inductance change as a function of target distance, which heuristically determined the chosen thickness for the modulating lattice layer.
Figure 3: Quantitative results demonstrating the co-design of mechanical and sensing properties by tuning lattice relative density. Inductance-force curves for (a) 7% , (b) 10% , and (c) 20% relative densities, illustrating how a higher relative density increases the operational force range at the cost of sensitivity (the slope of the curve). The data are presented as mean and confidence interval.
Figure 4: Experimental validation of the sensor’s robustness. (a) EM crosstalk when a linear vertical force is applied on S1 and the corresponding sensor response from its three neighbouring sensors. (b) Results from a 200-cycle mechanical repeatability test. The 99.96% overlap between the first three and last three cycles (inset) demonstrates the high durability and minimal plastic deformation of the 3D-printed gyroid structure.
Figure 5: Localized sensing and passive guidance of a manipulated object. The object is moved across the sensorized surface as it transitions from a low-stiffness region to a high-stiffness region, with its position tracked in a LabVIEW GUI. Two types of objects are demonstrated: Top Row: Rigid standard weight, Bottom Row: Irregular soft object
Figure 6: Object velocity and travel distance on tiles with varying relative densities. (a) The 20% region, being stiffer, exhibits lower friction, allowing continuous motion, whereas lower relative densities increase friction and cause faster deceleration. (b) Passive guidance of the object across regions with different relative densities. In the 20% region, the object maintains motion with minor velocity drops, while in the 7% region, it comes to rest rapidly due to higher friction.
Department of Electrical Engineering, Chalmers University of Technology, SE-412 96 Gothenburg, Sweden · Department of Automatic Control, Lund University, Sweden