1 / 22

FPGA implementation of population-based ant colony optimization

B. Scheuermann a ,∗, K. Sob, M. Guntsch a, M. Middendorf c, O. Diessel b, H. ElGindy b, H. Schmecka (2004) A review by Jacob Gillson. FPGA implementation of population-based ant colony optimization. Outline. P-ACO Introduction Algorithm Implementation on FPGA Experimental Results

jarvis
Télécharger la présentation

FPGA implementation of population-based ant colony optimization

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. B. Scheuermann a,∗, K. Sob, M. Guntsch a, M. Middendorf c, O. Diessel b, H. ElGindy b, H. Schmecka (2004) A review by Jacob Gillson FPGA implementation of population-based antcolony optimization

  2. Outline • P-ACO Introduction • Algorithm • Implementation on FPGA • Experimental Results • Contributions • Critique • Conclusion • Q&A

  3. Optimization • The goal is to find the best possible solution to a hard problem • Many problems take an impractical amount of time to solve exactly. Can be solved using heuristics • Combinatorial optimization attempts to find an optimal object (solution) from a finite set of objects • Travelling Salesman Problem • Minimum Spanning Tree

  4. P-ACO Overview • Population-basedAntColonyOptimization • Based on the path-finding abilities of ant workers • Uses a heuristic approach to solving combinational optimization problems • The ants use pheromone trails to guide other workers back to the food source

  5. Implementing P-ACO • Each FPGA maintains a colony of virtual “ants” in the form of solution generators • The generators choose their solutions based on pheromones made from past solutions • Each solution from each generator is compared to previous solutions • The most successful x solutions are added to a solution matrix

  6. P-ACO Architecture

  7. Finding Solutions • The pheromone matrix is initialized with random values • The ant chooses next path based on the pheromone strength at current node • Adds choice to a solution register • Repeats until all nodes have been used • Solution is compared to other ants • Best x solutions are added to population queue • New pheromone matrix is generated

  8. Choosing The Solution • The process continues until a stopping condition has been met • Set # of iterations • Best solution hasn’t changed after # iterations • The population module returns the elitist solution to the top module • This is hopefully a near optimal solution to the problem given

  9. Experimental Results • The colony was implemented on a Xilinx Virtex-II Pro • Compared to an AMD uni-processor system clocked to 1540 MHz • Runs a SMTTP (Single Machine Total Tardiness Problem) • Both the problem size n (number of nodes) and the colony size m (number of ants) are varied during experimentation

  10. Resource Use Uses a fixed number of ants m = 8 Uses a fixed problem size n = 64

  11. Resource Use • Due to increasing complexity, as the problem size and colony size increases, so must the on board resource use • With a fixed colony, the size complexity of LUTs and REGs increases at ~O(nlog(n)) • With fixed problem size, complexity is ~O(n)

  12. Time Comparisons

  13. Time Comparisons • The hardware implementation is at least 1.8x faster than software • Fixed problem size shows a large increase in speedup as all the ants run in parallel • Is limited by the number of BRAM cells on the FPGA • Fixed colony size shows a more modest speed-up • But only reliant on LUTs and REGs

  14. Contributions • Proves that P-ACO and other optimization algorithms can be implemented on commercially available FPGAs • Can be modified to obtain near-optimal solutions for most combinational optimization problems • Provides a large advantage over software-only implementations

  15. Critiques • Very in-depth and technical paper. Explains all decisions in detail and is entirely reproducible • Concepts become to technical at points and can be hard to follow • Many alternate designs and improvements suggested but were not acted upon • Hardware implementation is realistically limited to very small problem sets • The heuristic information was disregarded in the hardware implementation

  16. Conclusion • P-ACO algorithm successfully implemented on Virtex-II Pro FPGA • Showed that ant colony optimization was possible on commercial boards • Provided at least 1.8x speedup during experimentation over software • Limited to small problem sizes due to lack of resources

  17. References • Virtex-II datasheet -http://www.xilinx.com/support/documentation/data_sheets/ds083.pdf • Virtex-VII datasheet -http://www.xilinx.com/support/documentation/data_sheets/ds180_7Series_Overview.pdf • Combinatorial Optimization -https://en.wikipedia.org/wiki/Combinatorial_optimization • FPGA implementation of population-based ant colony optimization - B. Scheuermann et al. • Ant Colony Optimization - V. Maniezzo et al. • Ant diagram - https://en.wikipedia.org/wiki/Ant_colony_optimization

  18. Questions?

  19. Extra Slides

  20. Generator Block Architecture

  21. Population Block Architecture

  22. Comparator Block Architecture

More Related