A'R'M'S' Automated Robotic Messaging System

1
A.R.M.S.Automated Robotic Messaging System

• William Batts
• Chris Rericha

2
Objective

• Create a robotic vehicle capable of delivering
  papers and small packages throughout an office
  floor

3
Functional Overview

• Identify unique office locations
• Follow a predetermined paths between locations
• Ability to avoid reasonably sized objects
• Runtime user programmable destinations

4
Specifications

• Obstacle detection range 6 24
• Obstacle minimum size 4 wide x 8 high
• Maximum Number of Destinations 14
• Barcode length 5
• Minimum Obstacle/Barcode distance 2
• Maximum Package Weight/Size 1lb/8x12x1
• Minimum Runtime 5 minutes

5
Analytical Component

• IR Proximity Sensors
• IR rangers set in cross-fire configuration to
  provide proper width (IR ranger beam only 20º)

6
IR Ranger Calculations

• Given a maximum center-vehicle to IR beam
  distance of 6 and IR beam angle of gt20 degrees,
  proper placement and unprotected areas may be
  calculated.
• IR Ranger Placement Angle
• T arctan(Inner bound / Sensor-to-centerline)
  (Beam angle / 2)
• T arctan(6/3.5) (20/2)
• T 70 degrees

7
IR Ranger Calculations (contd)

• Unprotected Sides
• It is possible for a mobile object to move into
  an unprotected side area after the primary IR
  beam has passed
• Length of Unprotected Area
• L Distance from opposite IR ranger tan (T -
  10)
• L 7.5 tan(60)
• L 13
• Maximum Width of Unprotected Area
• W L tan (20/2)
• W 13 tan(10)
• W 2.3
• This blind spot only applies to mobile objects
  which will likely move into the IR beam before
  striking the vehicle

8
Opto-reflector Calculations

• QRD1114 (Shaft Encoding)
• Datasheet optimum range 1.27 millimeters
• Placement will be 1mm from wheel surface
• 1 barcode segments, 2.5 wheel, 31 shaft
  encodings minimum 3 samples per barcode
  segment (.25 per shaft encoding)
• QRB1114
• Datasheet optimum range 3.81 millimeters
• Placement will be 4mm from hard-flat surface
  (due to carpet pile sink)

9
Obstacle Avoidance

• Upon obstacle detection the vehicle will stop
• Re-examine path after a short period of time
• If obstacle remains, deviate from given path
• Intelligently determine avoidance solution

10
Path Sensing

• Front opto-reflectors will actively detect
  predetermined path edge
• When detected, vehicle will change track speed to
  move toward side of detection.
• Center opto-reflector serves as sanity check
  for path reacquisition during obstacle avoidance

11
Barcode Reading

• Opto-reflector used to read barcode
• Detect barcode header
• Move to middle of first bit (MSB) and record
  value
• Continue through all four bits of barcode

12
Hardware

• Fast Traxx Vehicle
• Motorola 68HCS12 Microprocessor
• Sharp GP2Y0D02YK IR Ranger
• TI SN75441 Quad Half H-Bridge DC Motor Driver
• Fairchild QRB1114 and QRD1114 Infrared
  Photosensors

13
Work Division

• William
• Opto-Reflector Testing
• Obstacle Avoidance
• Path Sensing
• Shaft Encoder
• DC Motor Interface
• Chassis Construction
• Component Testing

• Chris
• IR Ranger Testing
• User Interface
• Barcode Reading
• Program kernel
• PC Application
• System Test
• Component Testing

14
Work Completed

• Sensor testing and interface design
• Initial chassis construction and sensor placement
• Initial motor interface design
• Automatic barcode generation
• Microcontroller sensor test
• Preliminary software and UI design

15
Williams Work

• Barcode and Path Sensing
• Optimal distance to reflective surface is 4 mm
• Very reliable, rejects ambient light
• Shaft Encoders
• Optimal distance to reflective surface is 1 mm

16
Williams Work (Contd)

• Initial chassis construction

17
Williams Work (Contd)

• Initial DC motor interface design

18
Chris Work

• Sharp IR Rangers Testing
• Senses from 0.5 to 34
• Reliably to 24
• Detects objects in dim, halogen, and daylight
  equally successfully
• Detects objects of all colors in the visible
  spectrum equally well

19
Chris Work (Contd)
20
Chris Work (Contd)

• 24-bit bitmap barcodes automatically generated
• Sensor pin placement
• Sensor and system initialization code
• Microcontroller sensor test

21
Work to Complete

• Chassis and Sensor Mounting
• DC Motor to Microcontroller Integration
• Formal Algorithm Block Diagram and Implementation
• Software Kernel and System Initialization
• PC Application to Microcontroller Design
• User Interface Construction
• System Testing and Verification

22
Test Plan

• Test each sensor type for performance
• Develop a mini test for each component to verify
• Perform a code review for each software component
• Test basic sensor and movement ability on
  constructed vehicle

23
Test Plan (Contd)

• Test vehicle moving over barcodes at different
  possible angles of deviation
• Test vehicle on all different types of paths
• Test all combinations of obstacles and in varying
  daylight and obstacle colors
• Let complete system run for many hours through
  varying paths to work out performance bugs

24
Power Consumption
25
System Cost
26
Potential Safety Problems

• Vehicle may lose path and wander
• Remedied though search pattern (octagon) and
  timeout
• Kill button mounted for quick stop