Contact-rich robotic manipulation in unstructured environments demands reliable multimodal perception. Here, we present MagicGripper, a multimodal robotic gripper built around mini-MagicTac, a compact variant of the MagicTac sensor. Mini-MagicTac embeds multi-layer grid structures in a 3D-printed elastomer, enabling visual, proximity, and tactile sensing in a gripper-compatible form factor. In this paper, we introduce the design and multimodal perception capabilities of mini-MagicTac, as well as two algorithmic frameworks for proximity and contact detection. Experimental evaluations show that mini-MagicTac achieves high spatial resolution, accurate contact localisation, and robust force estimation under mechanical and manufacturing variations. Autonomous grasping trials further validate MagicGripper’s reliable multimodal perception and adaptability to complex manipulation scenarios. These results demonstrate MagicGripper as a compact and versatile platform for embodied intelligence in contact-rich environments.

In this paper, we present the design and benchmark of an innovative sensor, ViTacTip, which fulfills the demand for advanced multi-modal sensing in a compact design. A notable feature of ViTacTip is its transparent skin, which incorporates a ‘see-through-skin’ mechanism. This mechanism aims at capturing detailed object features upon contact, significantly improving both vision-based and proximity perception capabilities. In parallel, the biomimetic tips embedded in the sensor's skin are designed to amplify contact details, thus substantially augmenting tactile and derived force perception abilities. To demonstrate the multi-modal capabilities of ViTacTip, we developed a multi-task learning model that enables simultaneous recognition of hardness, material, and textures. To assess the functionality and validate the versatility of ViTacTip, we conducted extensive benchmarking experiments, including object recognition, contact point detection, pose regression, and grating identification. To facilitate seamless switching between various sensing modalities, we employed a Generative Adversarial Network (GAN)-based approach. This method enhances the applicability of the ViTacTip sensor across diverse environments by enabling cross-modality interpretation.