1 / 31

Course Outline

Course Outline. Preface Introduction to EHW Reconfigurable and Morphable Hardware Algorithms for self-configuration and evolution Demonstrations of Evolvable Systems Application Examples System Aspects Resources for EHW Engineers Summary. Introduction to EHW. Vision

ailis
Download Presentation

Course Outline

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Course Outline • Preface • Introduction to EHW • Reconfigurable and Morphable Hardware • Algorithms for self-configuration and evolution • Demonstrations of Evolvable Systems • Application Examples • System Aspects • Resources for EHW Engineers • Summary 1

  2. Introduction to EHW • Vision • EHW as Adaptive hardware • Components of a EHW system • Evolution of evolvable HW • Automated design and adaptation by search/optimization techniques • Evolution in Simulations and Evolution in Hardware • EHW for flexibility and survivability of autonomous systems 2

  3. Vision Deploy 1 a miniature 2 device in an unknown environment 3. Provide a high-level specification of intended function 4. The device adapts itself to provide the function intelligently5. 1 (drop, plug-in, etc); 2 (finger-nail size?); 3 (enemy field, remote planet, unknown computer); 4 (operation, mission); 5 (/optimally select/determine algorithms, protocols, resources to use, etc). 3

  4. Autonomous systems • Beyond physical reach to repair, beyond (or designed for absence of) communications • Increasing degree of autonomy • Unmanned Vehicles, UAV, UUV • Deployed sensing systems • Space systems • Places where humans can’t do it efficiently, economically, etc. Predator Global Hawk 4

  5. Autonomy in Space…the final frontier View of Venus Mars Interstellar exploration - Long life survivable spacecraft (100 years+) Space 2002 The 8th International Conference and Exposition on Engineering, Construction, Operations, and Business in Space co-located with Robotics 2002 The 5th International Conference and Exposition/Demonstration on Robotics for Challenging Situations and Environments. http://www.spaceandrobotics.org/index.html 5

  6. Global Interaction Medium (GIM)An evolving, symbiotic Omninet of 2050* • The HW/SW symbiosis will be morphing and adaptive • Interconnected info-processing systems could act as global brains • The GIM will become very closely interfaced with our sensing and thinking, so that we become ourselves the neurons of one or several brains. • Machines evolve, interconnections evolve... the result - a net whose power goes beyond anything that could be designed Internet Wired World Human Brains Energy resources Internet Wired World Service robots Living creatures Transportation systems Smart habitats “Only Connect: From swarms of smart dust to secure collaborative zones the Omninet comes to you”, by George Johnson, Wired Special Issue, Jan, 2000 6

  7. How Intelligent Technology Will Enhance Our World. Adaptation “Smart Quotient #1: Adapting 1 • The ability to adapt to users and the environment — to recognize context — is one of the basic attributes of smart systems. Can you plug your information device into a port and have it work, regardless of what device it is? Despite the underlying technical complexities, the goal of smart technology remains deceptively simple: to have computers adapt to us instead of the other way around. .. • Adaptive networks — In the future, communication networks will adapt instantly, organizing themselves into the most optimal configuration. Today, heavily used Web servers are mirrored on several sites around the world. When a user contacts the site by entering its URL, he or she is automatically directed to the most optimal site. The adaptive network makes this choice instantly, based on server usage and Internet traffic intensity. An important by-product of adaptive networks will be the ability to have a much better understanding of what is on the network because of so much intelligence in the network. • Technology Leadership articles from Get Smart, a report by CSC's Leading Edge Forum (LEF). http://www.csc.com/features/2001/25.shtml. Get Smart examines the role of "smart" technology in the marketplace; both now and in the years to come. 7

  8. Adapting • Adaptive interfaces — The way a system responds to its user by using available information is a telling sign of its adaptiveness. Many Web sites dynamically adapt to their visitors by using a technique called collaborative filtering — suggesting products or articles based on similar users' preferences. Amazon.com's book recommendations are a well-known example. • Adapting to location — If a system knows where someone or something is, it can better adapt to the current situation. Thanks to GPS (global positioning systems), people and cars with receivers can be located immediately by polling their communications device. For example, say you need to find the closest hospital. You turn on your PDA for directions. Its response depends on where you are: at home or in your car or in a distant city. • Adapting to system stress — When a system's functioning is endangered, it should be able to adapt and go on, without requiring a wholesale shut-down and abandonment of its users. Today's critical systems are backed-up by the "stronghold method" where another system takes over if the first one has a problem. Tomorrow's smart devices will use more sophisticated self-monitoring and self-healing techniques to adapt to stress and failure. For instance, a server under attack by a computer virus may detect the problem in an early stage and dynamically call upon an anti-virus provider for protection and a cure. • The more adaptive the system — whether in terms of networks, people, location or system condition — the better are its chances for long-term survival.” 8

  9. EHW as Adaptive hardware New users New functions Mismatches in fabrication Faults Adaptation Environments 9

  10. Adaptive Information Processing Adaptation can be at different levels: at architecture level there are polymorphic architectures that optimally re-combine heterogeneous resources, at lower level At device level devices digital (FPGA, reconfigurable DSP), analog and mixed signal (FPAA), NN Conventional (fixed) Devices (e.g CPUs) Adaptive Devices e.g. FPGA, FPAA, NN, Reconfigurable DSPs Polymorphic Architectures 10

  11. Environment Aware Devices Environment aware devices adapt to environment. For example when the battery is full they operate at high frequency and high resolution. If battery is low, frequency is reduced and resolution is reduced. For example an A/D converter that operates like this: If battery level is good – F=100MHz, Resolution = 16 Bit If battery is low – F=10Khz, Resolution is 8 bit 11

  12. Design Productivity Gap is Increasing Source: VLSI Technology Gates/cm2 Moore’s Law (59% CAGR) 3,830K 2,410K 1,520K >10M Gates +4 Power PC 957K Log Scale FPGA Capacity 603K 380K 305K 244K 195K 156K Average Cell-based Design Start (25% CAGR) 125K 100K 0.6µ 1994 0.5µ 1995 0.35µ 1996 0.25µ 1997 0.2µ 1998 0.15µ 1999 < 0.10µ 2003 Chips get bigger faster than our speed of doing designs. Even a single FPGA is now larger than an average design. 12

  13. A new generation of hardware A third generation hardware in terms of flexibility and fault tolerance Flexibility, fault-tolerance Self-reconfigurable, evolvable + Automated Design +Artificial/Computational Intelligence Reconfigurable 2005 -100nm - BISR, ITRS’99 Fixed HW Generation 1st 3rd 2nd 13

  14. Components of a Evolvable Hardware System HW that can change Mechanism controlling the change • Antennas EHW • search/optimization • algorithmic • knowledge-based • Electronics FPA +GA • MEMS • BioMEMS Evolvable Hardware = Reconfigurable HW + Reconfiguration Mechanism In a narrow sense (EHW) is programmable hardware self-configurable by built-in Evolutionary Algorithms. Same components for intelligent mixed-signal microsystems Flexible reconfigurable analog/mixed-signal devices intelligent part – the built-in mechanisms that would control the adaptation/self-configuration 14

  15. Evolutionary algorithms: inspiration from Nature “Design” goal: survival The most fit individuals survive becoming parents; children inherit parents characteristics, with some variations, and may perform better, increasing the level of adaptation. Evolution in nature has lead to species highly adapted to their environment: adaptation ensured survival. Millions of years Design goal: meet system specifications Potential designs compete; the best ones are slightly modified to search for even more suitable solutions. Same evolutionary principles can be applied to machines. Accelerated evolution, ~ seconds for electronics 15

  16. Design to be evolved The design to be evolved could be a program, model of hardware or the hardware itself Program Model of Hardware Physical Hardware 0 WhileTooFarFromWall 1 Do2 2 MoveForward 3 Do2 4 WhileInCoridorRange 5 TurnAwayFromClosestWall 6 WhileInCoridorRange 7 Do2 8 TurnParallelToClosestWall 9 MoveForward SPICE Netlist HDL code vdd 20 0 DC 5.0V vin+ 6 0 DC 2.5v m1 1 1 20 20 PMOS L={L1} W={W1} m2 3 1 2 20 PMOS L={L2} W={W2} Evolutionary is Revolutionary! 16

  17. Evolution http://www.oneonta.edu/~anthro/anth130/cartoons.html 17

  18. Evolution of EHW Evolvable Systems IP level Evolvable SOC We are here Programmable HW Downloadable SW Evolution of descriptions of electronic HW Board level EHW Evolution of computer programs Field Programmable Gate Arrays Evolutionary search for a parametric design Currently, the algorithms run outside the reconfigurable hardware; future solutions will be integrated System on a Chip and IP level Chip level 18

  19. Evolutionary synthesis and adaptation of electronic circuits Conversion to a circuit description Chromosomes • Evolutionary Algorithm • Search on a population of chromosomes • select the best designs from a population • reproduce with variation • iterate till goal is reached. 1011001101 0111010110 1101101101 Control bitstrings Models of circuits Extrinsic Simulators (e.g., SPICE) Evaluate responses, assess fitness Target response Intrinsic evolution Reconfigurable hardware Circuit response Potential electronic designs/implementations compete; the best ones are slightly modified to search for even more suitable solutions 19

  20. Extrinsic and intrinsic EHW Path from chromosome to behavior data file Parameters Model Simulator Reconfigurable HW Configuration Data file extrinsic Stimulus HW evaluator testing equipment Data file intrinsic 20

  21. EHW implementation: HW/SW • SWmodels RH/RM • SW • HW EHW = RH + RM • HW • Current approach to EH implementation: • Use RH- reassign cell function/interconnection • Use powerful parallel searches (e.g., GAs) to evolve the hardware In addition EHW requires • Fast evaluation • Low cost for failure Present solutions:RM in SW Future: everything seamlessly integrated in HW 21

  22. Evolution in Simulations vs Evolution in Hardware • Computationally intensive (640,000 individ. for ~1000 gen.) • 10s of hours, expected ~3 min in 2010 on desktop PC for experiments in the book (~50 nodes) • SPICE scales badly (time increases nonlinearly with as a function of nodes in netlist - in ~ subquadratic to quadratic way) • No existing hardware resources allow porting the technique to evolution directly in HW (and not sure will work in HW) • JPL’s VLSI chips allow evolution 4+ orders of magnitude faster than SPICE simulations on Pentium II 300 Pro. • ~ 10s of seconds in 2002 for circuits of complexity >= Koza’s). 22

  23. EHW vs NN AHW FPGA NN CNN EHW Inspiration • EHW seeks biological inspiration for • methodology leading to designs (1,2) • appropriate to situations/application • 1. of various types of HW • 2. freeing from biological constraints • NN seek biological inspiration for • computational elements, • architecture • mechanisms • for certain problems where biology does well (and attempts beyond) Mechanisms • Building block • NN: Simplified/distorted models of biological neuron • EHW: Domain oriented reconfigurable cell 23

  24. On-chip EHW vs CAD/synthesis tools Focus of our EHW work On-chip In-situ/in-field Autonomous synthesis Fully automated synthesis CAD EHW may overcome fabrication mismatches, drifts, temperature and other plagues to analog, exploiting the actual on-chip resources – finding a new circuit solution to the requirements with given constrains and actual on-chip resources. 24

  25. Evolware: from genetically engineered devices to evolvable space systems Mathematical-based search/optimization techniques Principles of natural evolution Evolutionary Algorithms Automated Design/Synthesis • CMOS circuits • Nano-electronic devices • Antennas • Proteins Self-adapting hardware • Reconfigurable HW continuously evolving in the target environment Evolvable Sensory Systems Evolvable HW/SW co-design Evolvable Robot Controls Evolvable Space Systems 25

  26. What kind of HW is needed for future missions Temperature & radiation tolerant electronics and long life survivability are key capabilities required for future NASA/JPL missions. Extreme environments - Temperature - Radiation Ultra long life Adaptive/Malleable Triton Venus 34.5 K 726K Autonomous 26

  27. EHW for flexibility and survivability of autonomous systems JPL/NASA driver – long–life spacecraft Dramatic changes in hardware/environment, e.g. in case of faults or need for new functions, may require in-situ synthesis of a totally new hardware configuration. Survivability: Maintain functionality coping with changes in HW characteristics Versatility: Create new functionality required by changes in requirements or environment EHW • Radiation impacts • Temperature variations • Aging • Malfunctions, etc. New functions required for new mission phase or opportunity Up-link new functions for re-planned mission Accurate model of hardware is not available after launch Develop space HW that can evolve 27

  28. Add evolvability to increase survivability • Long-Life Purposeful Survivability • really harsh/challenging environments • long-life (100+ years) • Evolvability • adaptation to environments • self-healing, self-repair • Advances in: • components • system robustness • space qualification • autonomy/intelligence 28

  29. How Evolvable Space Systems would Revolutionize NASA Missions • Long life, survivable, self-healing space systems • would allow long duration/far out missions • would harness required power and other resources from environment • Would enable evolvable missions capturing science/exploration • opportunities in real time • Space explorer • would produce knowledge from acquired data • would use the knowledge to mission refocus/replanning • would be able to create new functions, unforeseen before launch • would be able to learn on-the-fly to best deal with changing conditions • Fleet, Swarm, Armada • Salvaging: some do not adapt, their unharmed resources are reused by survivors Evolvable system technology: adaptive platform for space systems in a large variety of missions 29

  30. Ultimate goal: fully evolvable space systems • Morphing/plasticity can expand gradually from electronic subsystems to entire space systems. • Evolution of space systems would include autonomous changes/reconfiguration of both software and hardware: sensors, avionics, structure, ... • The future may see chameleon-like surface explorers and phoenix-like robotic birds... 30

  31. Fundamental open questions • Can we evolve artificial systems in similar ways natural systems evolve? Advantages and disadvantages. • How can we build devices/HW that evolve (autonomously)? • Can we seamlessly embed the guiding mechanism for evolution with the morphing system (i.e. the “goals”, the “goodness”) • How does EHW scale-up? • Can we use evolution to obtain intelligent systems, human competitive (and beyond) intelligence

More Related