Nano-to-micro integrated nonlinear components for smart autonomous aerospace systems

 

Salvatore Santoli

International Nanobiological Testbed (INT)

via A. Zotti 86, I-00121 Rome, Italy

e-mail intitsnt@uni.net

 

"Why can't we put spacecraft electronics on a chip? Perhaps we could use next gen-

                                                                                                                   eration materials. We could have smart structures. We could make spacecraft adaptive

                                                                                                               and autonomous, just like the human body, so we do not need hundreds of people on the  

                                                                                                                 ground watching dials and turning knobs. Advanced expert decision making tools along

                                                                                                  with nano- and microdevice technology, will open these possibilities."

Daniel Goldin

 

The technical limitations and costs of the classical methods for periodical maintenance of aerospace structures and for their control and performance, e.g. in the case of space missions which require a large human involvement, have made the "smart autonomous materials and systems" to appear as a promising concept. Moreover, in recent congresses the concept of a biomimicry-inspired technology has been advanced for realization of biomimetic chains of

sensing - information processing - actuating

which fully embody the concepts of "smartness" and "autonomy", either related to materials or to systems. Especially in the case of space exploration missions, extreme miniaturization has been promoted, mainly through the development of MicroElectroMechanical Systems (MEMSs) for decreasing the costs of probe launching, as mass and volume both scale as the third power of system size. This paper is devoted to show that nano-to-micro integrated systems, i.e. those featuring integration of the mesoscopic world with the macrophysical world, can be proposed to increase miniaturization and reach at least a quasi-biomimetic behaviour and even high-level biomimicry capabilities in information processing. The current efforts, concerning many usual materials and devices, for macroscopic level systems can be re-addressed toward a general nano-to-micro integration philosophy involving not only nano-MEMSs, but also material systems and devices other than micromechanical, i.e. electronic, optical and chemical. First the concept of "smartness" is analyzed, and the basic principles for nanostructured components of various nature to feature nonlinear, Hamiltonian and dissipative behaviour as computing elements according to proper design along the theoretical lines formulated recently are discussed. The problems met with in connecting mesoscopic to macroscopic physics for electronic nano-to-micro integration are then tackled; the novel architectures envisageable for nanostructured solid-state or molecular electronic systems and for sensing - data processing - actuating networks of a general kind are explored, mainly with reference to new aerospace applications, i.e. to the assembling or the inclusion, already in the manufacturing process, of nanostructured smart systems  into macroscopic members like preforms, prepregs, woven fabric, pultrusions, filament windings, embedded optical fibres, electroactive polymers, fullerene fibres etc. for active health monitoring, self-healing and active flight-control. Accordingly, smart multifunctional composites would be obtained, e.g. like an active, unidirectional-actuation 4-component laminate made up of 1) carbon fibre reinforced plastic containing multiply fractured optical fibres as the "nerves" and working as "bone" and "blood vessels", 2) an insulator epoxy resin, 3) aluminum as a "muscle", and 4) an electrode. It is argued that this nanobiomimetic technology means the bringing together of logic, physics and mechanical engineering, and a logic embodied through a hyper-interspersed nano/micro architecture  is proposed, consisting of  a peripheral autonomous level (e.g., a smart skin) made up of subsystems working on and interacting with the external reality just locally, in addition to a global action/motion level system depending on instructions controlled by a central unit. Small-size, finely interspersed sensor-processor-actuator units like thin film or microvolume bulk components, devices or systems, working e.g. on nonlinear/chaotic reaction-diffusion ongoings on artificial membranes, are shown to satisfy the requisites for a software capable of learning, as opposed to the rigid logic connections of standard hardware, and to give the possibility of processing signals from the environment without transforming them into a discrete form, i.e. into binary or ternary digits. Accordingly, the hyper-interspersed architecture, which is discussed through a "device - function" block diagram, is thought of as made up of single primitive computing units of nonlinear/chaotic and quantum holographic nature, characterized by specific transfer functions, interconnected into non-discrete pseudo-analogue networks performing computations decidedly different both from the digital and the analogue kind because their complex logic operations occur through energy pattern connections for recognition, decision making, learning, keeping attention, and computations involving both random and deterministic dynamics in a function dubbed "intermixing layer. Turning to a system-level overview, it is remarked that 1) the biomimicry-inspired nonlinear nanoscale-level energy/information relationships show an inter-dependence and inter-supplementability of the nondiscrete pseudoanalogue components and devices in the whole hardware/software system, from the mesoscopic up to the macroscopic scale; stated otherwise, computer function and its physical implementation may be closely connected throughout the whole hierarchy as a result of connection on the molecular level, against the current central paradigm of computer science, according to which computer concepts are independent of their physical realisations; 2) accordingly, the design should go "top-down" (i.e. from setting forth the system requirements down to the design of necessary subsystems, devices and components) as opposed to the current "bottom-up" approach, going the other way round; 3) compartmentalization down to the nanoscale level gives smartness as well as the mechanical strength and robustness necessary for the energy harvesting from the environment and the active shape control for flight, i.e. for two much eagerly sought after objectives for the next generations of flying systems.