1. Self-Reconfigurable Robot
(A Platform of Evolutionary Robotics)
O P Gujela
Vilnius Gediminas Technical University, Lithuania
opgujela@gmail.com

2. Outline
• Self-reconfigurable systems
• Modular transformer (M-TRAN)
• Demonstration of M-TRAN

3. Self-reconfigurable systems
• Artifacts based on homogenous modular
architecture
• Change their shape and function according
to the environment
(Self-reconfiguration)
• Able to assemble itself, and repair itself
without external help
(Self-Assembly, Self-Repair)

4. Homogeneous modular architecture
• The system made of many (mechanical)
modules
• Each module is identical in hardware and
software
• Each module has computational and
communication capability
• Each module can change local connectivity

5. Self-assembly and self-repair
Random shape Assemble target shape
Detect failure Cutting off Reassemble

6. 2-D Regular Tessellations

7. 2-D Self-reconfigurable hardware
Micro-module (MEL, 98)
Metamorphic robot (G.Chirikjian, JHU,93)
2-D Crystaline (M.Vona,
D.Rus, Dartmouth Col./MIT)

8. Fracta
Solid state module based on hexagonal lattice

9. Basic operations of fracta

10. Self-assembly
problem
How to change
connectivity among
modules to achieve
target configuration ?
Point to be Consider
• Modules are homogeneous
• Parallel and distributed
• Only local communication
• Physical constraints
Random
Given

11. Example: Self-assembly of fracta
11
Parallel algorithm based on connection
types and local communication
Connection types Target shape

12. 1. Each module evaluates
distance to the nearest
target configuration in
the program code
2. Modules compare the
evaluation through
simulated diffusion
3. Module which wins
among the neighbors
moves to random
direction
Parallel Distributed Self-Assembly
Type transition diagram

13. Difficulties in 3-D hardware
More mobility in limited space
• Spatial symmetry requires more degrees
of freedom
• More power/weight
• Mechanical stiffness

14. Space filling polyhedra
Rhombic
dodecahedron
Truncated
octahedro
n
Regular cube

15. Lattice based designs
3-D Crystaline
(M. Vona, D.Rus Dartmouth, MIT)
Design based on cube Design based on
rhombic
dodecahedron
Proteo M.Yim, PARC, 2000

16. Design based on cube
Molecule
Kotay, Rus, Dartmouth/MIT
3-D Universal Structure (MEL, 98)
Lattice based designs

17. Chain based designs
PolyBot: M.Yim ,Xerox PARC
CONRO: W-M.Shen, P.Will, USC

18. Lattice or chain ?
• Lattice based designs
• Reconfiguration is easy
• Motion generation is hard
• Requires many connectors & actuators
• Chain based designs
• Reconfiguration is hard
• Motion generation is easy
• Insufficient stiffness

19. M-TRAN Modular Transformer
 Hybrid of lattice and chain based
designs
• Easy self-reconfiguration and robotic
motion
• Two actuators
• Communication
• Stackable
• Battery driven

20. M-TRAN II

21. M-TRAN Module

22. Li-Ion battery
Power supply
circuit
SMA coil
Acceleration sensor
M-TRAN II
Neuron chipPIC
Main CPU
Connecting plate
Permanent magnet
Non-linear spring
Light bulb
PIC

23. M-TRAN I

24. New prototype
M-TRAN III Hook connection mechanism
• Quick
• Reliable

25. Coping with complexity
• Because of physical constraints such as
• Maintain connectivity
• Avoid collision
• Limited torque
• Non-isotropic geometry of M-TRAN module
makes self-reconfiguration very difficult
• Complexity can be relaxed by
• Automatic acquisition of rule set
• Heuristics (structured rule set)
• Periodical pattern in structure

26. Wall climbing
600 rules (no internal state)
Generated by software
18 rules (with internal state)
Hand-coded

27. Creeping carpet

28. Robot maker (structured rule set)

29. • Central Pattern Generator (CPG)
• Connected neural oscillators
• Oscillators entrain phases mutually
• Feedback of physical interaction
Rhythmic motion generation
CPG
Neural connection (CPG network)
Motor control Angle feedback
Mechanical interaction

30. CPG network
x
CPG
z
y
Excitatory
connection
Inhibitory
connection
Generate stable walk pattern (limit cycle)

31. CPG network tuned by GA
GA optimizes
• Connection matrix of
CPG
• Joint angles in initial
posture
by evaluating
• Energy consumption
per traveled distance
Simulation space
Given topology of robot
Initial set of individuals
Dynamics Simulation
Mutation, crossover
Selection
Download to modules
Yes
Generation +1
Converge

32. Thanks For Your
Kind Attention