Cascade LifeCycle: A Convenient Building Block for Efficient Robot Software Architectures

Software architectures dictate the organization of software within a robot. An effective architecture facilitates system scalability and adaptability for new tasks while managing critical aspects like computational load and real-time performance. This paper introduces the Cascade LifeCycle, a foundational component for software architectures. This component enables the construction of scalable and efficient architectures by cascading elements only when necessary to achieve the robot's active goals. This methodology supports the creation of efficient subsumption, three-layer architectures, or efficient subsystems, simplifying the programmer's task by concealing the complexity of component coordination. We present an implementation in ROS 2 utilizing LifeCycle Nodes and demonstrate our approach's effectiveness in robotic competition settings and university teaching environments using real robots.