Actuators are key components for moving and controlling a mechanism or system. However, the torque and efficiency of the current state-of-the-art actuators are insufficient and much lower than in humans. There are several applications (including prostheses, exoskeletons and running robots) where the unavailability of suitable actuators hinders the development of well-performing machines with capabilities comparable to a human. Remarkable, the power density and efficiency of electric motors are higher than a human muscle, so the problems of insufficient torque and efficiency resides in the transmission of the power and that the motors are not used at their highest efficiency. The first innovation of SPEAR is to solve the torque and efficiency problems, by investigating in depth a novel actuation paradigm, which I call Series-Parallel Elastic Actuation (SPEA) and that goes beyond variable impedance actuators. This new actuation paradigm is inspired by the series-parallel organisation of the muscle fibres. Modularity in actuation is currently introduced by placing in all joints the same motor, leading to over- or underactuated joints. In our body however, all the skeletal muscles are built of the same basic actuation unit: a muscle fibre. Modularity in actuation in a biological system is not at muscle level, but on a sublevel: the muscle fibre. SPEAR will introduce a second major innovation: the SPEA will introduce a basic actuation unit, a “transistor for actuation”. Such a SPEA-element is a missing link in robotics and will innovate the way robots are designed and built. The project will study the theoretical framework, the design principles, the control algorithms and the validation of demonstrators. SPEAR will fully answer all the research challenges and explore the frontiers of this novel actuation paradigm, leading to a tremendous impact on all engineered, actuated systems, especially in robotics.