FPGA Implementation of an Evolutionary Algorithm for Autonomous Unmanned Aerial Vehicle On-Board Path Planning

Kok, Jonathan; Gonzalez, Felipé; Kelson, Neil A.

doi:10.1109/tevc.2012.2192124

Cited by 71 publications

(37 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Kok et al [17] presents a novel solution which executes a developmental calculation for an equipment intended for unmanned elevated vehicle adjustment. While the study proves to be highly efficient when reaching the 10 Hz update frequency of a typical autopilot system, the number of individuals evaluated at once is limited to just 32.…”

Section: Genetic Programming On Fpgasmentioning

confidence: 99%

“…There has been great interest in applying reconfigurable solutions to genetic programming [5][6][7][8][9][10][11][12][13][14][15][16][17][18] and substantial progress has been achieved, however, there are still important limitations which restrict the applicability of these solutions in real environments and that we propose to address in our work such as: high latency due to fitness evaluation, simple trading strategies due to GPs represented with reduced complexity s.a. bit-strings instead of trees, small number of individuals evaluated, small number of iterations to reach the maturity of a GP population.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Run-time Reconfigurable Acceleration for Genetic Programming Fitness Evaluation in Trading Strategies

Funie

Grigoras

Burovskiy

et al. 2017

J Sign Process Syst

View full text Add to dashboard Cite

Genetic programming can be used to identify complex patterns in financial markets which may lead to more advanced trading strategies. However, the computationally intensive nature of genetic programming makes it difficult to apply to real world problems, particularly in realtime constrained scenarios. In this work we propose the use of Field Programmable Gate Array technology to accelerate the fitness evaluation step, one of the most computationally demanding operations in genetic programming. We propose to develop a fully-pipelined, mixed precision design using run-time reconfiguration to accelerate fitness evaluation. We show that run-time reconfiguration can reduce resource consumption by a factor of 2 compared to previous solutions on certain configurations. The proposed design is up to 22 times faster than an optimised, multithreaded software implementation while achieving comparable financial returns.

show abstract

Section: Genetic Programming On Fpgasmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Run-time Reconfigurable Acceleration for Genetic Programming Fitness Evaluation in Trading Strategies

Funie

Grigoras

Burovskiy

et al. 2017

J Sign Process Syst

View full text Add to dashboard Cite

show abstract

“…Then, the optimal path was calculated using genetic algorithms or PSO based on minimizing or maximizing the cost function. Also, researchers like Kok et al [13] applied executive points to their hardware routing implementation. In article [11] the environment was first examined and divided into different zones.…”

Section: Introductionmentioning

confidence: 99%

“…It utilized a combination of various constraints such as dangerous spots along the route distance and fuel consumption amount; Followedby the discovery of the optimal routes by using this performance function withthe aid of PSO and GA. Article [13] explained the implementation of a routing method on the hardware. The second section presents definition of the problemand our mathematical approach to solving the problem.…”

Section: Introductionmentioning

confidence: 99%

Offine/online Optimum Routing of a UAV using Auxiliary Points

Fadavi

Beheshti

Kermani

2017

IJECE

View full text Add to dashboard Cite

This paper presents a method to determine the route of a three-dimensional UAV. Three criteria; the height, the length of flight path and the un authorized areas are used as the constraints and combined in a fuzzy function as the evaluation function. The article aimed to discover a minimum cost route from source to destination considering the constrains. In this paper a new searching methodis proposed, with use of auxiliary points. The auxiliary point method iterativelydivides a straight line to two shorter lines with less cost of evaluation function. Implementation results show that the proposed method dramatically decreasesthe calculations; meanwhile the ight route is sub-optimum. Keyword:Evaluation Fuzzy systems Three-dimensional routing UAV Copyright © 2017 Institute of Advanced Engineering and Science.All rights reserved. Corresponding Author:Abbas Fadavi, Department of Mechatronics, Science and Research Branch, Islamic Azad Univercity, Semnan, Iran Email: abbas_fadavi@yahoo.com INTRODUCTIONUVA has its application mainly in the military industry [1][2][3]. However, it is used in a large number of non-civilian applications, such as fire fighting, meteorology, aerial photography and research projects. The aim of this article is to discover an appropriate route for the UAVs. An optimum route is a minimum distance route one which best satisfies the constraints. Different methods have been introduced and discussed in order to determine the optimal auxiliarypoints. Various articles have been presented to determine the optimal route forthe UVAs. A number of these articles have made use of classic methods. [11] applied hierarchical decision making. Their, the intended environment was first examined and divided into different areas. Then, according to the rate of the constraints satisfaction problems, the best identification areas and the paths were specified inorder to be able to reach the destination from the source among optimum zonesand to discover the best path. The method of Dijkstra was used. Robergeetal [12] presented a definite cost function. This function utilized a combination of different constraints like distance length with dangerous points and the rate of fuel consumption. Then, the optimal path was calculated using genetic algorithms or PSO based on minimizing or maximizing the cost function. Also, researchers like Kok et al [13] applied executive points to their hardware routing implementation. In article [11] the environment was first examined and divided into different zones. Then, optimal zones were identified. These are the zones which better satisfied the points inside the zones of

show abstract

“…Most of the time, achieving high-performance computation requires specific hardware that is able to sustain high MIPS/MFLOPS per watt or extreme high performance in very specific operations. Several technologies are appearing with promising results for high-performance computing on unmanned platforms, especially multicore architectures, but also FPGAs and GPUs [38,39]. The computation box also employs a general purpose "Main CPU" (at this moment Intel NUC [40]) in case a non-urgent task is required, limiting the energy consumption; or a high-performance "co-processor board" Multicore/FPGA/GPU in case high performance is required and the hardware is available.…”

mentioning

confidence: 99%

Hardware Design of a Small UAS Helicopter for Remote Sensing Operations

Royo

Pastor

Barrado

et al. 2017

Drones

View full text Add to dashboard Cite

This paper presents the hardware design and integration process employed to develop an Unmanned Aircraft System (UAS) helicopter. The design process evolves from the bare airframe (without any electronics), to become a complete and advanced UAS platform for remote sensing applications. The improvements, design decisions and justifications are described throughout the paper. Two airframes have been used during the design and integration process: the AF25B model and the more advanced AF30 model, from the Copterworks company. The airframe engine reliability and fuel economy have been improved by adding an Electronic Fuel Injection (EFI) and Capacitor Discharge Ignition (CDI), both managed by an Engine Control Unit (ECU). On-board power supply generation and regulation have also been designed and validated. Finally, the integration process incorporates on-board mission computation to improve the concept of operation in remote sensing applications. Several flight tests have been performed to verify the reliability of the whole system. The flight test results demonstrate the correct process of integration and the feasibility of the UAS.

show abstract

FPGA Implementation of an Evolutionary Algorithm for Autonomous Unmanned Aerial Vehicle On-Board Path Planning

Cited by 71 publications

References 22 publications

Run-time Reconfigurable Acceleration for Genetic Programming Fitness Evaluation in Trading Strategies

Run-time Reconfigurable Acceleration for Genetic Programming Fitness Evaluation in Trading Strategies

Offine/online Optimum Routing of a UAV using Auxiliary Points

Hardware Design of a Small UAS Helicopter for Remote Sensing Operations

Contact Info

Product

Resources

About