X-Hinges Software for 3D Printed Sensors: Design Guide

What is X-Hinges software for 3D printing?

X-Hinges is a new design software from MIT that allows you to create 3D-printed objects that perceive and adapt to human touch. By using this tool, you can design items like lamps and gloves that feature automatically embedded sensors.

The release is currently a preprint or press release and has not undergone peer review.

How does embedded sensor technology solve signal interference?

Researchers at MIT’s Computer Science and Artificial Intelligence Laboratory and Tianjin University developed X-Hinges to address signal interference in interactive 3D-printed devices. In earlier versions, both sensors and wire-like traces were made from the same material, which produced similar, hard-to-interpret signals.

X-Hinges fixes this by automatically embedding two distinct materials during the printing process. One material is designed to send a strong sensor signal, while the other is a wire-like element that avoids interference.

What can you design with X-Hinges?

You can select up to three axes of movement for your designs: compressing, moving up and down, or moving to the sides. The software provides four interface shapes for you to choose from: a custom option, cylinder, rectangle, and thin outline.

Researchers have used this to create a glove that records how finger joints bend to teach a robot how to grasp blocks. They also created an origami-style lamp that senses when you pinch it to adjust its brightness.

You can also build devices that identify objects placed on them, such as recognizing a banana based on tactile data. For a more custom approach, researchers built a video game controller that looks like a cartoon shark by using a custom shape and setting the tail to move up and down.

How does the software make objects responsive?

The software makes objects responsive by embedding sensing capabilities that allow them to adapt to your interaction. Jiaji Li explains that this adds a conscious element to everyday items, similar to how an AI agent is programmed to make decisions based on your instructions.

Xiang Chang notes that while most methods only allow for one axis of motion due to signal interference, X-Hinges enables your item to move from three angles. The software also records motion data in each of those areas at the same time, which makes the device responsive to different kinds of interactions.

What are the current limitations and challenges?

X-Hinges faces a calibration bottleneck because 3D-printed electrical elements vary slightly. This means that two devices appearing identical from the outside may not provide the same sensor readings.

To address this, Chang and Li built a computer vision system that can recalibrate outputs to ensure data is accurate across two of the same items. They hope to develop a built-in signal recalibration mechanism for the software in the future.

Key takeaways from MIT’s 3D‑printed sensor research

The research team behind X-Hinges includes lead author Xiang Chang, a PhD student at TJU and visiting researcher at CSAIL, and senior author Jiaji Li, an MIT EECS postdoc and CSAIL researcher. The paper, titled “X-Hinges: 3D-Printed Self-Sensing Compliant Mechanisms for Continuous and Multi-DOF Motion Sensing,” was also written by MIT EECS associate professor and CSAIL principal investigator Stefanie Mueller and recent TJU graduate Haiyang Yan.

The project received support from the MIT-GIST Program and a postdoctoral research fellowship from Zhejiang University. You can see the team present their work at the ACM Symposium on User Interface Software and Technology in early November.

What is the future of this research?

Jiaji Li’s research focuses on letting sensing and actuation capabilities grow directly inside a 3D-printed object, rather than adding them on after the printing process. This builds on his previous work with printed tendon-driven mechanisms that gave objects the ability to move.

His next goal is to combine both actuation and sensing in a single print. Li states this will create new physical platforms for AI agents and embodied intelligence.