SuckTac: Camera-based Tactile Sucker for Unstructured Surface Perception and Interaction

TL;DR

The paper tackles the challenge of endowing robotic suction cups with high-fidelity perception on unstructured surfaces. It introduces SuckTac, a camera-based tactile sucker embedded via a multi-material casting process, and jointly optimizes its geometry (cross-section, corrugated lip, and microstructures) to enhance both sensing and adhesion. Through analytical membrane modeling, texture- and roughness-perception experiments, and integrated perception–action demonstrations on a robotic arm and a soft robot, the work shows robust, texture-aware grasping and locomotion capabilities. This approach expands the frontiers of intelligent suckers, enabling more reliable manipulation and exploration in diverse, real-world environments.

Abstract

Suckers are significant for robots in picking, transferring, manipulation and locomotion on diverse surfaces. However, most of the existing suckers lack high-fidelity perceptual and tactile sensing, which impedes them from resolving the fine-grained geometric features and interaction status of the target surface. This limits their robust performance with irregular objects and in complex, unstructured environments. Inspired by the adaptive structure and high-performance sensory capabilities of cephalopod suckers, in this paper, we propose a novel, intelligent sucker, named SuckTac, that integrates a camera-based tactile sensor directly within its optimized structure to provide high-density perception and robust suction. Specifically, through joint structure design and optimization and based on a multi-material integrated casting technique, a camera and light source are embedded into the sucker, which enables in-situ, high-density perception of fine details like surface shape, texture and roughness. To further enhance robustness and adaptability, the sucker's mechanical design is also optimized by refining its profile, adding a compliant lip, and incorporating surface microstructure. Extensive experiments, including challenging tasks such as robotic cloth manipulation and soft mobile robot inspection, demonstrate the superior performance and broad applicability of the proposed system.

Figures (6)

• Figure 1: Bionic camera-based tactile sucker and its biological inspiration and robotic applications: (a) Octopus utilize suckers to feel the surface condition of the target; (b) Enlarged view of the natural sucker structure; (c) The proposed bionic camera-based tactile sucker can support robot arm to classify the object and picking as (d), and assist the path planing for soft robot crawling in (e).
• Figure 2: The detailed Structure of SuckTac. The camera-based tactile sensor is artfully integrated into the sucker (Right: Exploded view).
• Figure 3: Characterization and performance evaluation of the SuckTac: (a) Modeling of the sucker. (Left) Modeling parameters; (Right) comparison between theoretical predictions and experimental results.(b) Suction force–displacement relationships of four different sucker structures. (c) Suction force comparison of suckers with four lip designs on four types of periodic curved surfaces. (d) Suction force of suckers with different sub-millimeter structure densities tested on sandpapers of varying grit number.
• Figure 4: SuckTac perception experiments: (a) Surface texture samples of 18 daily objects acquired by the SuckTac. (b) Confusion matrix of texture classification using a ResNet18 network. (c) Differential images of sandpaper surfaces (top row) and the corresponding frequency spectra (bottom row).
• Figure 5: Grasping-classification experiments with the SuckTac: (a) Experimental setup and three leather surfaces of distinct textures. (b) System framework integrating vision, robotic arm control, and pneumatic actuation. (c) Workflow of the grasping-classification procedure.