Rible-Sim/Rible.jl

Rible.jl is an academic prototype addressing the niche of rigid-flexible multibody simulation. While differentiable simulation is a high-value field for robotics (allowing for gradient-based hardware and control optimization), this specific project shows almost zero adoption, with only 3 stars and 2 forks over nearly three years. It likely serves as a code accompaniment to a specific research paper rather than a maintained software project. In the Julia ecosystem, it is overshadowed by more mature projects like RigidBodyDynamics.jl. On a broader scale, it competes with industry-standard differentiable simulators such as Google's Brax (JAX-based) or NVIDIA's Warp, which offer significantly better performance, hardware acceleration, and community support. The defensibility is minimal because the implementation is relatively straightforward for anyone familiar with Julia's autodiff ecosystem (Zygote/Enzyme) and standard multibody dynamics. Frontier labs are unlikely to compete directly as this is too domain-specific, but the project is highly vulnerable to obsolescence as more general-purpose simulators incorporate flexible-body support.