Hybrid planning computes task plans in full observability (a sequence of possible actuation actions) via the integration of low-level feasibility checks in classical planning approaches. Conditional planning extends the classical planning framework to account for plans under partial observability. We use a hybrid conditional planning approach to compute plans that include sensing and actuation actions in solving real-world problems by addressing uncertainties due to incomplete information at the planning phase. We generate a hybrid conditional plan of actions using parallel instances of a non-monotonic hybrid \hcplan\ that supports defaults, integration of external computations, and non-deterministic actions. We show our hybrid conditional planning framework's \hcplan\ applicability through service robotics scenarios with manipulation and navigation tasks. Furthermore, we evaluate the effect of parallel threads on the computation of hybrid conditional framework on different benchmark scenarios for the service robotics domains.