Logistics involves the storage and displacement of goods. These goods are stored in warehouses and shipped to retailers in pallets. Pallets are produced on a customer’s order. Order picking for a pallet is a fatigue induced process resulting in poor performance of the workers, decreasing the productivity and inducing delays in supply chain. Flexibility is introduced to increase productivity by commissioning robots for process automation. These robots consist of autonomous ground vehicles for transporting freight and static manipulators for pick and place. A static robot has limited workspace and the capability of a manipulator is significantly enhanced by adding a mobile base. Mobile manipulation is now being exploited for pick & place and pallet production. This thesis presents a first attempt to achieve autonomous palletization using mobile manipulation. To acquire palletization by mobile manipulation requires the identification of functional blocks, to conceive a framework to achieve this task. A thorough state of the art has been prepared in this thesis corresponding to each element of the global framework. To realize the proof of concept, a prototype has been developed by leveraging existing technologies, by integrating a mobile base with manipulator and a grasping system with a gripping element. For each functional block of the global framework, control execution strategies have been developed and tested in industrial environment. Specifically, localization is acquired by the use of synthetic landmarks, a motion planning and control strategy is employed for global navigation and a rack tracking motion control has been developed for moving inside the racks. To combine and execute all the elements without deadlocks a coordination framework is used as a global supervisor. The path planner for global navigation is based on the shortest distance between two points, and rack tracking is developed by applying the conventional Hough transform to the lidar data and using the output in a nonlinear controller, while the motion planner for manipulation is based on linear trajectories. The framework for supervisory control is based on discrete event systems topology and state machines corresponding to each element have been modelized using Petri nets. Finally, the framework has been tested for a complete picking task on the mobile manipulator to validate the selection of strategies and performance of each functional element. The successful demonstration has been concluded as a first step towards the evolution of autonomous palletization.