LEARNING, SELF-DIAGNOSIS AND MULTI-OBJECTIVE CONTROL OF AN ACTIVE TENSEGRITY STRUCTURE
Bernard Adam and Ian F. C. Smith
Ecole Polytechnique Federate de Lausanne (EPFL),
Applied Computing and Mechanics Laboratory,
Station 18, GC-G1-507,
CH-1015 Lausanne, Switzerland
E-mail: Ian. Smith@epfl. ch
This paper presents a full-scale active tensegrity structure at EPFL and demonstrates how it can leam as well as carry out self-diagnosis and self-compensation. Tensegrities are generally flexible structures: small loads may lead to large displacements. We thus control slope by actively modifying the self-stress state between cables and struts. The structure benefits from past experience through case-based reasoning. It memorizes past control commands and adapts them in order to react to new applied loads up to forty times more rapidly than without this previous control information. Redundancy of this structure provides opportunities for “fault tolerant” behavior. The active control system can also be used to perform self-diagnosis and then to self-compensate local damage. For many cases of local damage, the structure remains capable of satisfying control goals. This paper also summarizes a multi-objective optimization method for control according to four criteria. In contrast with other applications involving multiple objectives, such as design where users prefer choices, this is a control task, thereby requiring identification of a single solution only. Also, the single dominant objective usually generates hundreds of possible solutions. Four objectives are evaluated firstly using Pareto optimality and then a unique solution is chosen through successive filtering of candidate solutions using a hierarchy of objectives. The combination of advanced computing techniques with structural control of serviceability criteria is providing many new possibilities for structural engineers. These results are expected lead toward more autonomous and self-adaptive structures that are able to evolve as their environment changes.
Tensegrities are spatial and lightweight structures composed of compressed struts and tensioned cables that are stabilized by a self stress state. They are very flexible; small loads can induce large deflections. Serviceability control, performed by modifying the self stress state of the structure, has potential to create opportunities for using this type of structures in practical applications.
The active tensegrity structure built at EPFL contains five modules and covers a surface area of 15m2. It rests on three supports that altogether block six degrees of freedom in three dimensions. Two theses have thus far been completed using this structure by Fest (2002) and Domer (2003).
Each module consists of twenty-four stainless steel cables and six composite fiber bars that are connected to each other through thirteen joints. Compressed struts converge toward a central node and this constitutes the particularity of this design as inspired by the office of Passera & Pedretti, Lugano (Switzerland). The central node reduces the buckling length of compressed elements and leads to more slender struts (Fest et al, 2003).
The structure is equipped with three displacement sensors (nodes 37, 43, 48) and ten actuators: see Figures 1 and 2. This makes it possible to actively control the structure. The actuators are placed longitudinally in in-line pairs within each module and this makes it possible to modify the self stress state through modifying the strut length. Control commands (sequences of active strut contractions and elongations) are identified using stochastic search (Domer et al, 2006). This is one of the first large-
M. Pandey et al. (eds), Advances in Engineering Structures, Mechanics & Construction, 439-448.
© 2006 Springer. Printed in the Netherlands.
scale active tensegrity structures that are able to satisfy a serviceability criterion. Djouadi et al. (1998), Skelton et al (2000) and Sultan (1999) studied tensegrity structure control only through numerical simulation.
Figure 1: EPFL tensegrity structure equipped with 10 actuators and 3 displacement sensors
The active control system can also be used for system identification if members break. Actuators make it possible to perturb the structure while sensors measure the response. Local damage does not necessary lead to a total collapse since the structure is redundant. The active control system is then used to self-compensate a broken element and satisfy the serviceability criterion considering a certain loss of carrying capacity.