Control Barrier Function Toolbox: An Extensible Framework for Provable Safety

The need for safety is ubiquitous, however, guaranteeing safety can be difficult for systems with non-trivial dynamics. Control barrier functions (CBFs) are an active area of research for safety-critical control systems. Generally, the application of CBFs has been limited to those with the controls expertise required to write the safety constraints from scratch. An analogous technique for stabilization, Control Lyapunov functions (CLFs), can be paired with CBFs to find safe and stabilizing controls. Our CBF Toolbox, written in Python, enables easy construction of CBF and CLF constraints to provide safety guarantees in simulation and hardware demonstrations. Additionally, the CBF Toolbox serves as a useful tool to teach the theory, and explore the impact of CBFs and CLFs on control systems. We discuss the basic theory and organization of the toolbox, a simple software example, and a robotics demonstration to land a quadrotor on a moving platform with the CBF Toolbox.