This paper describes a novel architecture inspired from the multicellular organizations found in Nature. This architecture is tailored to let functional organisms (logical functions) grow on silicon. To this aim, the silicon surface is populated with an array of identical programmable cells, which may be configured by a bitstream. By analogy with the biological world, the concatenation of the bitstreams used to program the cells composing a given function is called the “genome” of that function. In addition to conventional BIST (Built-in Self-Test) structures addressing signal line faults, this new version tolerates failures affecting power supply. It also allows the growth of differentiated organisms on the same surface by including a code in the genome to distinguish them. As a testbed, we have developped an integrated circuit prototype, code name GenomIC. It contains only a single 4-cell structure, but prefigures which kind of structure can be massively integrated in very large circuits in order to manage complexity (multicellular organization), evolvability (genetic data manipulation) as well as fault tolerance.

A 128 x 128 pixels, 120 dB vision sensor extracting at the pixel level the contrast magnitude and direction of local image features is used to implement a lane tracking system. The contrast representation (relative change of illumination) delivered by the sensor is independent of the illumination level. Together with the high dynamic range of the sensor, it ensures a very stable image feature representation even with high spatial and temporal inhomogeneities of the illumination. Dispatching off chip image feature is done according to the contrast magnitude, prioritizing features with high contrast magnitude. This allows to reduce drastically the amount of data transmitted out of the chip, hence the processing power required for sub-sequent processing stages. To compensate for the low fill factor (9%) of the sensor, micro-lenses have been deposited which increase the sensitivity by a factor of 5, corresponding to an equivalent of 2000 ASA. An algorithm exploiting the contrast representation output by the vision sensor has been developed to estimate the position of a vehicle relative to the road markings. The algorithm first detects the road markings based on the contrast direction map. Then, it performs quadratic fits on selected kernel of 3 by 3 pixels to achieve sub-pixel accuracy on the estimation of the lane marking positions. The resulting precision on the estimation of the vehicle lateral position is I cm. The algorithm performs efficiently under a wide variety of environmental conditions, including night and rainy conditions.

Computation in Cellular and Molecular Biological Systems, pp. 223-235 (1996) No AccessGENOMIC CELLULAR AUTOMATA TRANSPOSED ON SILICON: EXPERIMENTS IN SYNTHETIC LIFEP. MARCHAL, P. NUSSBAUM, C. PIGUET, S. DURAND, D. MANGE, E. SANCHEZ, A. STAUFFER and G. TEMPESTIP. MARCHALCSEM Centre Suisse d'Electronique et de Microtechnique SA, Jaquet-Droz 1, CH-2007 Neuchâtel, Switzerland, P. NUSSBAUMCSEM Centre Suisse d'Electronique et de Microtechnique SA, Jaquet-Droz 1, CH-2007 Neuchâtel, Switzerland, C. PIGUETCSEM Centre Suisse d'Electronique et de Microtechnique SA, Jaquet-Droz 1, CH-2007 Neuchâtel, Switzerland, S. DURANDCSEM Centre Suisse d'Electronique et de Microtechnique SA, Jaquet-Droz 1, CH-2007 Neuchâtel, Switzerland, D. MANGELaboratoire de Systèmes Logiques, Swiss Institute of Technology, Ecublens, CH-1015 Lausanne, Switzerland, E. SANCHEZLaboratoire de Systèmes Logiques, Swiss Institute of Technology, Ecublens, CH-1015 Lausanne, Switzerland, A. STAUFFERLaboratoire de Systèmes Logiques, Swiss Institute of Technology, Ecublens, CH-1015 Lausanne, Switzerland and G. TEMPESTILaboratoire de Systèmes Logiques, Swiss Institute of Technology, Ecublens, CH-1015 Lausanne, Switzerlandhttps://doi.org/10.1142/9789812819826_0017Cited by:2 PreviousNext AboutSectionsPDF/EPUB ToolsAdd to favoritesDownload CitationsTrack CitationsRecommend to Library ShareShare onFacebookTwitterLinked InRedditEmail Abstract: The following sections are included: Introduction GCA Architecture: Stacked Layers of "Proto-Cells" Preliminary Definitions Top Layer CA: Coordinate Computation Mid-Layer CA: Gene Selection out of the Genome Memory Bottom Layer CA: What Functional Proto-Cell? Emergence of Lifelike Properties Cell Reproduction (Cloning) Cell Differentiaiion (Organism Realization) System/Organism Differentiation Healing (Local Self-Repair or Cicatrization) System/Organism Healing (Global Self-Repair) Method of Construction: an Example Conclusion Back to von Neumann Specialist Viewpoints Future Developments References FiguresReferencesRelatedDetailsCited By 2Ontogenetic hardwareMoshe Sipper, Daniel Mange and André Stauffer1 Dec 1997 | Biosystems, Vol. 44, No. 3Functional organisms growing on siliconPascal Nussbaum, Pierre Marchal and Christian Piguet8 June 2005 Computation in Cellular and Molecular Biological SystemsMetrics History PDF download

A novel architecture descending from the work of von Neumann, has been developed. This architecture borrows its main principles from living systems. Like living beings, the organisms considered here are able to autonomously develop, maintain their functionality and reproduce. These genomic architectures are developed on reprogrammable hardware. They are not restricted to a given class of functions but accept any combinational and sequential function to be downloaded. These architectures are fault tolerant by design, so they can adapt to failures affecting the silicon. They autonomously evolve so as to maintain their functionality and hence self-reconfigure when needed.

A method for the graphical specification and the automatic generation of analogue behavioural models is presented. This method has been implemented as a new software tool called ABSynth. The behaviour of the component to model is described as a functional diagram, which is then automatically translated into a VHDL-A-like analogue hardware description language. No syntax knowledge is necessary and the modelling time is reduced.