Abstract

We show how the limited electrical power and FPGA compute resources available in a swarm of small UAVs can be shared by moving FPGA tasks from one UAV to another. A software and hardware infrastructure that supports the mobility of embedded FPGA applications on a single FPGA chip and across a group of networked FPGA chips is an integral part of the work described here. It is shown how to allocate a single FPGA's resources at run time and to share a single device through the use of application checkpointing, a memory controller, and an on-chip run-time reconfigurable network. A prototype distributed operating system is described for managing mobile applications across the swarmbased on the contents of a fuzzy rule base. It can move applications between UAVs in order to equalize power use or to enable the continuous replenishment of fully fueled planes into the swarm.

Highlights

  • The term swarm is usually identified with a group of living organisms who arrange themselves to cooperate to achieve a common task that no one of them could complete as an individual

  • field programmable gate array (FPGA) are an appropriate platform for small UAVs because they have low power requirements yet can compute high complexity tasks such as image processing

  • We have shown that mobility of FPGA tasks between members of the swarm will allow a swarm to be active for a longer period of time or allow the continuous replacement of members of the swarm

Read more

Summary

INTRODUCTION

The term swarm is usually identified with a group of living organisms who arrange themselves to cooperate to achieve a common task that no one of them could complete as an individual. UAVs that cooperate to achieve a common task (such as geolocation) in an autonomous way (using agents) have been given by analogy the title of swarm in this paper. This is the first time an operating system for reconfigurable computing has been implemented to execute practical embedded applications.

PREVIOUS WORK AND REVIEW OF LITERATURE
Capabilities and applications of single small UAVs
FPGAs as compute platforms for small UAVs
Advantages of swarms of UAVs
Sharing resources in a swarm: a typical scenario
Agents mobility and mobile agents
Conclusion
Introduction
Basic elements of an embedded operating system for FPGAs
Dynamic partial reconfiguration and real FPGAs
Checkpointing
Sharing resources amongst applications
Memory network
Memory allocation and arbitration policies
Implementation of resource arbitration
Experience running the applications under the OS
SHARING FPGA COMPUTING AND POWER RESOURCES ACROSS A SWARM OF UAVs
Using agents for resource sharing
The UAV swarm agent environment
Migration rules
Findings
CONCLUSION
Full Text
Paper version not known

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call

Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.