OpenExo Opens the Door to DIY Exoskeletons

There are certain fields that are, for one reason or another, very difficult to break into. In the world of artificial intelligence, for instance, developing frontier models is out of reach for most people due to the massive amount of computational power necessary for the training process. Space exploration is also a difficult area for experimentation due to the high costs of getting payloads into orbit, and the specialized hardware that is required to withstand harsh environments.

Until very recently, exoskeleton technology also fell into this category. These wearable devices are capable of restoring independence to those with movement disorders, preventing injuries in manual laborers, and giving superhuman physical abilities to everyone. But to date, expertise in mechanical and software engineering, physiology, electronics, and artificial intelligence has been a requirement for designing an exoskeleton. Furthermore, the technologies involved in these systems tend to be very complex and expensive.

That has limited the number of people that can experiment with exoskeletons, which in turn has slowed forward progress. That is not good for anyone, so a team at Northern Arizona University and the University of Michigan has developed what they call OpenExo. It is an open source modular exoskeleton development platform that was built to remove the traditional barriers to entry in the field. In particular, it minimizes the areas in which developers need expertise, and reduces the length of a typical development cycle.

OpenExo is the first comprehensive open source framework for developing robotic exoskeletons, and it provides everything a developer might need, including detailed design files, code, and build guides. The modular nature of the system means it can be used to build single-joint or multi-joint devices for various body parts, including the hips, ankles, and elbows. The project’s creators hope this flexibility will enable a new generation of researchers to explore diverse use cases, from mobility assistance to weightlifting support.

The system utilizes inexpensive and user-friendly hardware platforms like the Teensy 4.1 and the Arduino Nano 33 BLE development boards to control the motors and sensors. This not only brings costs down significantly, but also makes OpenExo-based exoskeletons much easier to build and program.

Most prior exoskeleton systems were locked into specialized lab setups and tailored to narrow applications, making them difficult to adapt and reproduce. This rigidity discouraged new researchers and contributed to what some have called a reproducibility crisis in the field. OpenExo addresses this by offering a flexible architecture and standardized interface, enabling others to adapt the platform to new problems without starting from scratch.

The team has already used their technology to help children with cerebral palsy walk more comfortably and to assist patients with gait disorders in rehabilitation. They have also demonstrated the versatility of the system across a range of configurations: hip-only incline assistance, ankle-only indoor and outdoor support, combined hip-and-ankle load carriage, and elbow-based weightlifting assistance.

With the release of OpenExo, the researchers hope to see a wave of innovation that makes robotic exoskeletons more common, more affordable, and more impactful. The code is available in a GitHub repository, and extensive documentation is available, for those that want to experiment with the framework for themselves.