Safety is essential for robots in unknown environments, especially when there is physical Human-Robot Interaction (pHRI). Control over energy, or passivity, is an effective safety mechanism. However, when the control algorithm is implemented in a discrete-time computer, computation and communication delays readily lead to loss of passivity and to instability. In this paper, a way to make the actuators aware of the energy that they inject into the system is presented. Passivity and stability are then always guaranteed, even in situations of total communication loss. These Embedded Energy-Aware Actuators are a model-free passivity and safety layer that make complex robotic systems dependable, well-behaved and safe. The proposed method is validated in simulation and experiments.

In the field of aerial robotics, one of the key challenges is to enable aerial manipulators to exert substantial forces on the environment. Enabling this will allow the technology to perform meaningful tasks airborne, such as cleaning or grinding surfaces. While in contact and applying a large, continuous force, control of the UAV's attitude is a challenge. In this work, we show that a regular (PID-based) attitude controller is incapable of stabilizing aerial manipulators that apply physical contact forces on the environment that are comparable to the UAV's weight. We present a novel control algorithm that uses an LQR-optimized state feedback on the roll and yaw angle while in contact. Experiments on a UAV of 1.5 kg show that the proposed controller is capable of applying a contact force of over 15 N - equal to the UAV's weight - sustained for several minutes.

Quadrupedal running is an efficient form of locomotion found in nature, which serves as an inspiration for robotics. We believe that a resonance-based approach is the path towards energy-efficient legged locomotion and running robots. The first step in working towards this goal is creating an energy-efficient one-legged hopper. Such a one-legged hopper was designed and constructed. The impact efficiency of the mechanism is calculated analytically, determined in simulation and measured with the prototype. The impact efficiency as calculated from the experiments is found to be in agreement with the analytical expectation and simulation results. Finally, using an electric motor to inject energy by creating a virtual spring with reverse hysteresis, hopping is achieved.

Synchronisation is an essential part of many controlled dynamical systems, in particular in the limb motion of legged robots. In this paper we introduce a novel control strategy that allows synchronisation of two oscillators without using any external power, but by modulating the power flow between the two oscillators. We then derive a separate energy-level controller that regulates the oscillation amplitude, again without changing the system's total energy. Finally, we show that the strategy works on a realistic mechanical system, by synchronising the phase difference and apex height of two bouncing masses.

Legged locomotion involves periodic negative and positive work, which usually results in high power consumption. Improvement of the energy efficiency is possible by using energy storage elements to reversibly store the negative work performed during a walking or running cycle. While series elastics with high impedance (high gear ratio) actuators are widely used, we investigate the application of parallel stiffness with highly backdriveable actuators. We specifically show that the use of parallel springs in a bounding quadruped with a flexible spine can lower power consumption by over 50%.