Robotic Hands is a field of biomimetic robotics that draws a huge amount of attention from the scientific community and relevant advances in the mechanical, control and electronic areas. A major problem in the design of these systems is the great number of components that must be controlled at the same time and in some cases at high speed. This problem is often approached using a grid of microcontrollers, which are easy to use and program, nevertheless can sometimes become complicated to deal with, especially when there is a large number of controlled components. Another solution is to explore the intrinsic parallelism using FPGA (Field Programmable Gate-Arrays) allowing the implementation of hardware architectures achieving realtime requirements. Nonetheless, system development in FPGAs demands more time and effort from the designer being hard to debug. In this scenario programmable SoCs (System on Chips) combine flexibility and robustness making possible to exploit the advantages of both hardware and software implementations. This work presents a novel approach in the control of a multi-DoF (Degree of Freedom) robotic hand using a SoC, specifically a ZYNQ-7010 chip, in order to implement a modular force control scheme used to perform grasping tasks. The proposed system combines the use of an ARM microprocessor to synchronize the functions of several custom logic blocks. The proposed solution was tested in a real robotic hand and the performance from the computational point of view was contrasted against other implementations such as an ATMEL micro-controller, and a full-dedicated hardware solution.