Robotic Lunar Exploration Program

1
Robotic Lunar Exploration Program
2
Overview1.1 Project Objective1.2 Lunar
Surface Reference Missions Proposed Tasks
3
1.1 Project Objective

• Build a robotic prototype, controlled
  autonomously if possible, that will address
  operational issues. 
• It should be able to manipulate and collect soil
  samples, and dig trenches. 
• It should have easy adaptability for unpredicted
  instruments, repair work, part replacement, and
  all around versatility. 
• It will benefit space exploration by helping
  build a lunar habitat for astronauts to live and
  train for other missions, a refueling station
  for ships, as well as future unforeseen projects.

4
1.2 LSRM

• Scientific exploration.
• Determining suitability of Moon
• Developing technology and conducting tests
  relevant to long-term human stays
• Developing tests relevant to further explorations
  i.e. Mars and beyond
• Testing technologies that might lead to economic
  benefits
• Understanding crew health, safety and
  performance, and effectiveness for long stays on
  the moon

5
2. Research 2.1 Non-prehensile Manipulation 2.2
Excavation Methods 2.3 Microcontrollers 2.4
Other Electrical Research
6
2.1 Non-prehensile manipulation

• Grasp
• Kinematic grasp
• Fixturing
• Quasistatic grasp
• Using sliding and stationary forces

• Static grasp
• Positioned by gravity
• Dynamic grasp
• Positioned using accelerations

7
Nonprehensile Palmar Manipulation with a Mobile
RobotBy W.Huang and G.Holden from RPI

• Huang and Holden used a palmar robot with two
  degrees of freedom to pick up a box sitting on
  the floor pressed against a wall.
• They used a Kinematic analysis to model the
  interaction between the box and the palm of the
  robot.

8
Non-prehensile conclusions

• Focus on grasp-less manipulation because it
  requires less actuators, allows more degrees of
  freedom, and versatility is increased overall.
• The downside is that planning, controllability,
  and programming become more complicated.

9
2.2 Excavation Methods

• Building a robotic machine to accomplish this
  task before astronauts arrive will save money and
  reduce risks involved.
• A backhoe-type autonomous robot is one idea that
  could be used for many other tasks as well

• In order to set up a permanent habitat on the
  moon, trenches will need to be dug in order to
  cover important cables, protecting them from
  machinery and radiation from the sun.

10
Wheel Trench Cutter

• A wheel with lots of scoops
• In front, side, or behind
• Many sizes of scoops

11
Bucket Excavator

• Hydraulic arm with scoop attached to the end
• Very popular

12
Chain Trencher

• Similar to Wheel trencher
• Uses steel teeth attached to a chain

13
Vibratory Plows

• Creates a duct by vertically vibrating a plow
  blade.
• Cable is laid into the duct

14
Excavation Conclusions

• Bucket excavators are versatile but extremely
  difficult to control.
• Trenchers/vibratory plow are very task specific
  and do not offer much versatility.

15
The Hardware Big Picture
2.3 Microcontrollers
Power Supply
Communication
Actuators
Computational
Sensors
16
Distributed Computational System
Sensor
Sensor
Small controller
Small controller, (i.e.. Basic Stamp)
Actor
Actor
Main Controller (i.e.. Rabbit)
Sensor
Small controller
Sensor
Small controller
Actor
Actor
17
Microcontroller Conclusions

• Distributed calculation system is better than one
  strong and central system
• Redundancy, no real-time operating system
  necessary and available and well supported
  hardware available at the university
• Use the microcontrollers available at the
  university
• Supporting persons available, controllers are
  cheap and fit our needs, specialized controller
  would not bring advantages that would over come
  increased cost

18
2.4 Other Electrical Research

• Autonomous Robotics
• Robot Vision
• Navigation
• Inverse Kinematics
• Robotic Control Systems
• Decision Making Systems
• USB Interface

19
3. Design Phase 3.1 Target Specifications 3.2
Tasks 3.3 Learning Tools 3.4 Field Experience
3.5 Soil Force Test
20
3.1 Target Specifications
Rank Description Category Spec Unit
1 Trench depth Excavation 6-10 in
1 Trench length Excavation 10 m
1 Soil sample size Collection 3 ft3
2 Object manipulation Strength 10 Kg
2 Operating time Power 1 task W
21
3.2 Tasks

• Top 5 in Bold
• Soil Sample Collection
• Trenching
• Cable Laying
• Rock Chipping
• Soil Compaction
• Dislodging
• Rock Flipping
• Path Clearing
• Soil Relocating
• Rock Pushing
• Soil Sifting
• Tilling
• Stake Planting

22
3.3 Learning Tools

• Robotic arm
• Used to test and visualize motion and
  controllability of proposed design
• Computer interface will allow us to test
  controllability using a programmable set of
  commands

23
3.3 Learning Tools

• Boe Bot
• Will give us experience working with continuous
  motion servos, microcontrollers, and
  programmability of each
• Options to set up our own circuit for testing of
  sensors and other motors

24
3.4 Field Experience

• We took a trip to Northern Idaho to get some
  hands-on experience with a back-hoe.
• We learned that hydraulics are powerful, but
  would be a controls nightmare due to their
  nonlinear operation.
• The Kinematics of an arm with four degrees of
  freedom would be too difficult to try to program
  in a one-year project.

25
Field Experience
26
Soil Force Test
27
4. Prototypes 4.1 Decision Process 4.2 Hole Saw
Arm 4.3 Poker/Drill 4.4 Poker Arm 4.5 Sewing
Machine 4.6 Sandlot Arm
28
4.1 Decision Process
Prototype Decision Matrix Prototype Decision Matrix Prototype Decision Matrix
Aaron Jen Matt Jason Victor Total
Vic's all-n-one 6 7 7 5 8 33
Arm with poker 7 12 11 9 7 46
Bucket 6 10 11 9 6 42
Hole Saw/drill 10 7 9 10 8 44
Vibratory plow 7 9 7 6 9 38
Sewing machine 9 10 10 7 8 44
Sandlot/Spatula 6 9 11 11 9 46
Conveyor belt 5 8 8 9 6 36
Chain Trencher 8 7 7 5 7 34

29
4.2 Hole Saw
30
4.3 Poker/Drill
31
4.4 Poker Arm
32
4.5 Sewing Machine
33
4.6 Sandlot Arm
34
5. Initial Program Scripts 5.1 Flip a rock 5.2
Chip a rock
35
5.1 Flip a Rock

• This script uses the Sandlot Arm
• Get to a rock
• Send out ultrasonic signal
• Receive back ultrasonic signal
• Use algorithm to calculate distance from robot to
  rock
• Lower Sandlot Arm to ground
• Use linear motor to extend arm to the rock
• Use ultrasonic signal and motor encoder signals
  to verify arm is at the rock
• If arm is at rock, continue if arm isnt close
  enough move closer and repeat verification
• Raise arm slightly off ground
• Rotate arm 180
• Use camera to take picture of flipped rock

36
5.2 Chip a Rock

• This script uses the Arm Poker
• Get to a rock
• Send out ultrasonic signal
• Receive back ultrasonic signal
• Use algorithm to calculate distance to rock
• Position Poker Arm so that the poker is above
  rock
• Use linear motor to strike down and chip the rock
• Repeat strike motion once to make sure rock is
  chipped
• Use a camera to take a picture of chipped rock.

37
6. Budget

• Available...7500
• Research and learning1000
• Prototypes and testing....1000
• Mechanical components and machining..2000
• Electrical components.1500
• Other.500