It’s no surprise that Skydio, with its impressive AI and autonomy capabilities is making waves in the robotics community. The company has been sharing learnings from their motion planner system to help others build robots at scale for years now through SymForce – a library of pre-trained computer vision models available on GitHub so everyone can contribute!
Skydio’s autonomy team was able to develop SymForce in five years, thanks largely due the company’s focus on performance and code maintainability.
SymForce makes it possible to code a problem once, experiment with it symbolically, and then generate optimized code that solves the problem.
The code generation library not only makes it possible to model complex robotics problems as symbolic expressions but also adds components like 3D geometry types, camera models and noise algorithms.
Skydio’s VP of Autonomy, Hayk Martiros says thatSymForce is an “incredible tool” for rapid prototyping and running highly optimized runtime code on SkyDrones. He goes onto explain why he made the decision to make it open source with GitHub:
This general-purpose workflow is effective for solving a wide variety of tasks in robotics and related domains, and can speed up common tasks by an order of magnitude while requiring less handwritten code and reducing the surface area for bugs.
We think its core components can be useful in any domain that requires algorithmic code and can benefit a wide range of audiences from high-school students to tech companies.
SymForce is a powerful robot software that is being used by tens of thousands of robots at Skydio. The public library is new and in beta stage. You can install SymForce using pip, play around with it in a notebook, and deploy production code in a few hours. You can also head over to the Skydio blog to learn more about how SymForce works and why it is so fast.