In this chapter, we present a distributed, hierarchical locomotion control strategy designed for whole-body cooperation and capable of scaling to large groups of agents. Our approach leverages a hierarchical structure to decompose complex tasks into smaller, more manageable subtasks. By incorporating spatiotemporal continuity features, we establish the sequential logic essential for causal inference and cooperative behavior in tasks that unfold over time. This structure enables efficient and coordinated control strategies. Through training within this framework, we demonstrate enhanced adaptability and cooperation, resulting in superior task completion performance compared to traditional methods. Additionally, we introduce a set of environments as a benchmark for evaluating embodied cooperation.