Search papers, labs, and topics across Lattice.
This paper introduces PRISM, a comprehensive multimodal dataset designed to enhance robotic learning in industrial settings by capturing high-precision, contact-rich manipulation tasks. Unlike existing datasets that focus on simpler pick-and-place operations, PRISM encompasses over 25 complex tasks, including electronic component manipulation and conveyor sorting, with a total of 45 hours of teleoperated demonstrations. The dataset's rich multimodal data鈥攊ntegrating RGB-D, force/torque, tactile, and robot-state measurements鈥攑rovides a robust benchmark for advancing multimodal perception and control in real-world manufacturing scenarios.
PRISM reveals that high-precision industrial manipulation requires a dataset that captures the complexities of contact-rich tasks, challenging the limitations of existing datasets focused on simpler operations.
Recent progress in robotic learning has been fueled by large-scale datasets collected in everyday environments. However, most existing datasets emphasize short-horizon, low-contact tasks such as pick-and-place, and therefore do not capture the precision control, force/torque or tactile regulation, and multimodal feedback required for industrial assembly. To address this gap, we introduce PRISM, a large-scale multimodal dataset for contact-rich industrial operations. The dataset spans more than 25 manipulation tasks (e.g., electronic components plug/unplug, conveyor-based sorting) and covers diverse mechanical constraints. PRISM includes more than 5,000 trajectories totaling 45 hours of teleoperated demonstrations, recorded using synchronized multi-view RGB-D, force/torque, tactile, and robot-state measurements. In contrast to datasets collected in household or laboratory settings, PRISM provides a realistic benchmark for multimodal perception and control under high-precision industrial constraints, and serves as a foundation for contact-rich, generalizable manipulation in real-world manufacturing environments. The dataset is open-sourced at: https://tengbo-yu.github.io/PRISM/
