On Three-Layer Architecture

1
On Three-Layer Architecture

• Erann Gat
• Jet Propulsion Laboratory
• California Institute of Technology

Presentation by Ekkasit Tiamkaew Date 09/09/04
2
Historical Background

• Mid 1980s SPA architecture
• Sense-Plan-Act.
• Then came subsumption architecture
• Repudiation of plans.
• Composing all layers by means of overriding.
• More task-specific
• E.g. Herbert. Successful but unreliable.

3
SPA Shortcomings

• Planning and world modeling are difficult
• Planning is time-consuming.
• Lose sync with reality
• Unexpected outcome can cause subsequent plan
  steps to be executed in an inappropriate context.
• Running Researcher Syndrome

4
Shortcomings of Reactive

• Not sufficiently modular.
• Quick solution wire elimination.
• Heavily rely on sensor mechanism
• (mirror-like) reflections cause fault readings.
• Lack mechanisms to manage complexity
• Sometimes internal states are necessary

5
Alternate Approaches

• 1989-1990 Tooth Rocky III were built
• More reliable.
• Subsumption architecture with higher layers of
  computational abstraction.
• 3-component architecture
• Reactive feedback control system
• Deliberative planner
• Sequencing mechanism
• e.g ATLANTIS 3T

6
Internal States in 3 layers

• Stateless in the controller
• (or skill layer in 3T)
• Past memory in the sequencer
• (or sequencing layer in 3T)
• Future prediction in the deliberator
• (or planning layer in 3T)

7
The Controller

• Primitive behaviors wall-following,
  collision-avoiding, etc.
• Compute a function with constant-bounded time
  space complexity
• Able to detect a failure
• Avoid using internal states
• Ok for ephemeral states.

8
The Sequencer

• Select a primitive behavior to be done at a given
  time under each situation (state)
• State enumeration
• No robots execution history is used.
• Sometimes not possible to know its state.
• Conditional sequencing
• Respond to contingencies.
• Manage multiple parallel interacting tasks.

9
The Deliberator

• Time-consuming computations
• e.g. Planning, search-based algorithms, etc.
• Either
• 1) produce plans for the sequencer or
• 2) respond to specific queries from the sequencer

10
Robot Alfred

• Controller runs on an electronic control board
• Sequencer Deliberator run on a notebook
• Multiple sonar sensors, each is 15 degree rotated
  from its neighbors.

11
Control Layer in Alfred

• obstacle avoidance
• Obstacle in soft-left, not in soft-right
• Obstacle in soft-left, not in soft-right
• Obstacle in front

12
Control Layer in Alfred (cont)

• wall-finding
• Turn towards the sonar with the shortest range
  reading.
• Move forward until an obstacle is in hard region.
• wall-alignment
• Turn slowly until a discontinuity is seen by the
  forward sensor.

13
Control Layer in Alfred (cont)

• wall-following
• Keep track of the robots heading odometer
  reading.
• If a sonar reading is much greater than the last
  known distance, correct the course.
• Stop if travels more than 2 meters without valid
  reading

14
Sequencing Deliberative Layers in Alfred

• With an obstacle in front, first turn one way,
  then the other, slowly increasing the angle
• Determine its location, move to the rooms
  center, head towards one of 3 doors
• Locate a wall, verify and follow it, trying match
  the path with the shape of the maze (deliberative
  layer)
• Plan the path (deliberative layer)

15
Conclusions

• Planning alone is insufficient, but not
  unnecessary
• Reactive approach is another extreme idea towards
  an opposite direction
• No One True Architecture. Mix them!
• 3-layer architecture still ignores issues like
  sensor processing and world modeling, which may
  be needed in the future.