The diffusion of Human-Robot Collaborative cells is prevented by some barriers. Classical control approaches seem not yet fully suitable for facing variability conveyed by the presence of human operators beside robots. Heterogeneous knowledge representation capabilities and abstract reasoning are crucial to enhance flexibility of control solutions. This work presents SOHO (Sharework Ontology for Human Robot Collaboration), a novel ontology specifically designed for Human-Robot Collaboration. The paper describes the pursued context-based approach, the novelty of the designed ontology with respect to the state of the art and shows its validity in a realistic Human-Robot Collaboration scenario.