In this paper we introduce an analysis of dependability of an elementary yet critical component of robotic systems designed to operate in environments shared with humans, i.e., the joint-level actuation system. We consider robot joints that implement the variable impedance actuation (VIA) paradigm. The VIA has been demonstrated to be an effective mean to achieve high performance while constantly keeping injury risks to humans by accidental impacts below a given threshold. The paper describe possible implementations of the VIA concept which use the Antagonistic Actuation (AA) in three different arrangements. This study follows a previously reported paper dealing with safety. Here a detailed comparative dependability and performability analysis in front of possible specific failure modes is conducted, whose results provide additional and useful guidelines for design of safe and dependable actuation systems for physical human-robot interaction.

This paper is concerned with the design and control of actuators for machines and robots physically interacting with humans, implementing criteria established in our previous work [1] on optimal mechanical-control co-design for intrinsically safe, yet performant machines. In our Variable Impedance Actuation (VIA) approach, actuators control in real-time both the reference position and the mechanical impedance of the moving parts in the machine in such a way to optimize performance while intrinsically guaranteeing safety. In this paper we describe an implementation of such concepts, consisting of a novel electromechanical Variable Stiffness Actuation (VSA) motor. The design and the functioning principle of the VSA are reported, along with the analysis of its dynamic behavior. A novel scheme for feedback control of this device is presented, along with experimental results showing performance and safety of a one-link arm actuated by the VSA motor.

In this paper, an independent joint position and stiffness adaptive control for a robot arm actuated by McKibben artificial muscles is reported. In particular, muscular and dynamic parameters of the system are supposed unknown. Adaptive control performance is tested in a one degree of freedom experimental setup and compared with PID control performance. The adaptive control scheme is then applied to a robot arm that is conceived to perform tasks in anthropic environments. The adaptive control developed is such that performance of the robot arm is very similar to human arm performance. Experimental results are reported.

We describe some initial results of a project aiming at the development of a programmable compliance, inherently safe robot arm for applications in anthropic environments. In order to obtain safety in spite of worst-case situations (such as unexpected delays in teleoperation, or even controller failure), we propose an approach to achieving the compliance by mechanical rather than by control design. We first describe some of the control problems encountered in a typical, large, possibly unknown mechanical compliance, and present the result that shows the possibility to cope with these uncertainties in an adaptive way. Next, we describe the initial development of a new prototype arm under construction in our laboratory. The arm is designed to achieve arbitrary position tracking in 3D with controlled effective compliance at the joints.