ReadingResearchRadarInvestment frameworkAI Company Primary Sources
Sign in / Sign up
← Back to reading
NewsMaterial3 min readDisconfirmAI Editorial

Columbia University Professor Li Yunchu: Assembling in 3 Minutes with 3 Hyperparameters to Scale Tactile Sensing Like Vision

At IROS 2026, Li Yunchu introduced the core design logic of FlexiTac, an open-source tactile sensor: by choosing a sufficiently simple piezoresistive mechanism, simulation requires only a spring-damper model and 3 hyperparameters to achieve high-fidelity alignment with real signals. Assembly takes just 3 minutes, ensuring batch-to-batch signal consistency. This strategy lowers the barrier to entry for tactile sensors, enabling undergraduates to integrate them using 3D printers, which has fostered an ecosystem covering multimodal policies, world models, and VLA feedback within a year and a half of being open-sourced.

2026-10-30 16:13 GMT+8

Robotic vision has established a relatively unified system of sensors and data, but tactile sensing remains an un-converged weak link in embodied intelligence systems. At the tactile session of IROS 2026, Li Yunchu, a professor at Columbia University (United States), pointed out that different laboratories often use fully self-developed, mutually incompatible tactile sensors. These sensors collectively face the dilemma of struggling to balance lifespan, flexibility, multi-dimensional information perception, resolution, cost, and consistency simultaneously. This makes tactile data collection, algorithm research, and reproducibility across different systems significantly more difficult.

The strategy proposed by Li Yunchu's team is not to pursue more complex sensors, but to deliberately choose a sufficiently simple sensing mechanism. FlexiTac is an open-source flexible piezoresistive tactile sensor featuring a layered structure of Flexible Printed Circuits (FPC) and force-sensitive films. The team has also open-sourced the complete Bill of Materials (BOM), detailed assembly instructions, and software for reading data. Its core design logic is: because the sensing mechanism is simple, simulation implementation only requires building a spring-damper system and calibrating 3 hyperparameters to achieve high-fidelity alignment between simulated tactile signals and real-world signals.

Sign in / Sign up
ReadingResearchRadarInvestment frameworkAI Company Primary Sources

This simplification brings chain reactions. In terms of assembly, once materials are prepared, it takes only 3 minutes to complete; in terms of form, the sensor is like a sheet of paper that can be bent, cut, and wrapped around various irregular curved surfaces, adapting to gloves, fingertips, or robot skin. Regarding consistency, Li Yunchu cited an example: his student Binghao made four sensors in the summer of 2024 to record videos. A year and a half later, using this same set of sensors, a model trained on data collected a year ago still worked perfectly. Output signals also showed good consistency across different batches, addressing a common pain point for many tactile sensors containing rubber materials.

Simulation adaptability is the key hub of this strategy. Li Yunchu's approach involves first collecting random interaction data in the real environment, then replaying the same actions in the simulation environment to collect two sets of tactile data and plot histograms. The calibration goal is to align these two distributions. He emphasized: the more precisely simulation aligns with reality, the less randomization is needed in simulation, making sim-to-real transfer more accurate. The same framework has been validated in laboratory dual-arm assembly tasks and on NVIDIA humanoid robots.

In scaling real-world data, FlexiTac fills a gap in the UMI handheld teleoperation device. The original UMI only recorded precise pose tracking information, ignoring the tactile feedback provided by internal mechanical springs to human operators. By integrating FlexiTac, the tactile feedback felt by human operators can be indirectly recorded as sensor output signals, allowing collected data to contain both visual and tactile information synchronized in time.

Since being open-sourced about a year and a half ago, FlexiTac has gained widespread adoption in academia and industry, spawning work in multimodal policies, deformable world models, quadruped locomotion-manipulation, and VLA tactile feedback. The VTAP Gripper project, a collaboration with Purdue University (United States) on multi-finger dexterous manipulation, was shortlisted for the Best Paper Award at this IROS. Analog Devices (United States) is advancing its commercial version, which achieves five times the tactile resolution of a human finger and is mass-produced based on semiconductor manufacturing processes.

However, Li Yunchu acknowledged that FlexiTac is just one of many tactile solutions, and the field has yet to form a unified standard. During the Q&A, he confirmed that piezoresistive materials cannot directly measure shear forces; they can only approximate shear force estimation when spatial and temporal resolutions are sufficiently high. This means that for operational tasks requiring precise tangential force judgment, the current version of FlexiTac still has inherent limitations.

Li Yunchu summarized the goal of this work as 'democratizing tactile perception': starting from a simple sensing mechanism developed over decades, he aims to promote sensors to more researchers and scale up tactile perception through two paths: simulation and real physical environments.

Sources

  • leiphone.com↗
  • leiphone.com↗
Continue researching →