Emotional agents are becoming promising technologies for real-time control applications (i.e., robotics). The principal controllers of the agent are the emotional processes that decide the selection of behaviors to fulfill the objectives. The number of emotional processes increases with the complexity of the application, limiting the processing capacity of a mono-core processor to solve complex problems. A costly solution would be the use of HPC computers to solve complex problems. However, the possibilities of parallelization of the emotional processes permit their execution in parallel on multicore processors, enabling the agent to solve problems of higher complexity at a low cost. This paper presents the implementation of the emotional processes of a robotic agent in a multicore processor. To this end, the parallel emotional processes are identified and characterized, and a real-time system based on the EDF scheduling policy, to execute the emotional workload on the multicore, is proposed.In the experiments, mobile robotic applications are set-up taking into account different environmental conditions, robot dynamics and emotional states. The applications are run on multicore multithreaded processors depending on the workload requirements of the applications. Results show that only a threecore with 6 threads can tackle complex problems in the worst emotional conditions, a dual-core solves less constrained problems under normal emotional conditions and a single-core processor is only suitable for simple problems under the best conditions.