SelfAssembly, Monitoring, Checking, and Steering of MAV Sensing Grids

1
Self-Assembly, Monitoring, Checking, and Steering
of MAV Sensing Grids
Diana Gordon and William Spears Naval Research
Laboratory Insup Lee and Oleg Sokolsky
University of Pennsylvania Tugkan Batu, Ronitt
Rubinfeld, and Patrick White Cornell University
2
Objective

• To combine
• Artificial Physics framework
• Monitoring and Checking framework
• Program checking
• to solve an important Navy problem.

3
MAVs Scenario
strike and counter- measures MAVs
surveillance MAVs
model checking, theorem proving, abstraction
subplan
strike and counter- measures MAVs
global plan
subplan, data
updates that require changes in plan
crypto protocols
subplan
program checking and monitoring
threat analysis
4
MAVs Scenario Detailed View
Global monitoring UAV
MAVs form a hexagonal lattice sensing grid
Hexagonal sensing grid suggested by Rick Foch,
NRL MAV expert
5
Constructing a Hexagonal Lattice
R
6
Local Rules for Self-Assembly in the Artificial
Physics Framework

• Force law
• F G m m / r
• F repulsive if r lt R else attractive

2
1
2
7
More Features of the Artificial Physics Approach

• Fault-tolerance via redundancy at nodes.
• Flexibility to self-assemble into a variety of
  lattice types.
• Offline evaluation measure of lattice quality
  using angular error.
• Local self-repair mechanisms.

8
Need for Online Global Monitoring and Checking
with Steering for Repair

• Need an online measure of lattice quality not
  dependent on measuring angles.
• Need global steering (repair) methods that
  restore lattice quality after severe disturbances.

9
Vulnerability of MAVs to Air Turbulence

• Size lt 6 in length, width, and height.
• Speed 20 - 40 mph.
• Weight 50 - 70 grams.
• Low Reynolds number (directly proportional to
  size, speed, and weight of vehicle)
• Inertia can be ignored only viscosity of air
  relevant.

10
Potentially Hazardous Sources of Air Turbulence

• Wind shear
• Severe updraft
• An explosion from a bomb

11
Severe Disturbance to Lattice
Global monitoring UAV
12
Need for Online Program Checking

• Internal and external conditions potentially
  damaging to MAV processors.

13
Vulnerability of On-board MAV Processors

• Potential intra-vehicle hazards due to
• Proximity of parts because of small vehicle size.
• Potential environmental hazards due to
• Unexpected weather conditions,
• Low flying altitude increases risk of exposure
  to radiation.

14
Particular Hazards That Could Damage an MAV
Processor

• EMI from an insufficiently shielded motor close
  to the processor, or from power lines.
• Extreme temperature variations.
• Ocean spray or wet weather can cause corrosion.

15
Our Solution

• Include MaC program checking.
• Our combined approach is the following...

16
Global Monitoring of Formation

• UAV or other global observer monitors (without
  seeing the pattern) the progress of the MAV
  formation into a lattice
• Check if increasing number of alerts (an alert
  occurs if an MAVs neighbor is at the wrong
  distance from it)
• If signif. increase in alerts, issue a global
  alarm
• If global alarm, command a global parameter
  change for steering (lattice repair), e.g.,
• Change gravitational constant G, or
• Temporarily suspend repulsive force between MAVs
  (done in demo)

17
Program Checking to Detect Processor Failure

• Global monitor checks the MAVs math
  co-processors
• Monitor tests the MAVs trig methods based on
  their I/O.
• If failure, broadcast to MAVs to emulate the trig
  functions in software.
• Monitor also self-tests its own trig methods.

18
Conclusions

• Contribution
• Synthesis of
• Artificial Physics framework
• Monitoring and Checking framework
• Program checking
• solves an important Navy problem.
• See demo!