Navigation names, definitions, and classes as agents, their skills and dependencies
Roger 17 November
First, some thoughts about the names of classes. • Pilot was probably not the best choice. MovmentController might be better; shorter versions include MovementControl, MoveController, Controller . Any of these would be better than PoseController because pose is the dynamic state of a navigating robot and it changes as the result of a movement. It cannot be controlled directly. While there is a couple of years of code written by many users that would be broken if we abandon Pilot, if we ever going to do it, release 1.0 is the time. But I am inclined to let it stand. • I don’t care for Qualified as the QualifiedCoordinates in microedition.location, but prefer Probabilistic
Definitions of terms • Navigation: control of movement in the plane. • Directions in a the plane are relative to the direction of the X axis ; the direction of the Y axis is 90 degrees. • Heading: the direction the robot travels while moving forward. (if it moves backwards, the direction of motion is the heading + 180 degrees) • Pose: the current coordinates of the robot and its heading In probabilistic robotics, the pose is a random variable, the value of which is never known exactly. But decisions are based on a belief about the pose. This belief is represented as a probability distribution, the two most popular representations being the multi-variate normal and particle set. In this context, I use the term pose to mean “belief about the pose” • Elemental movements are : • travel in a straight line – change coordinates but not heading • rotate in place – change heading but not coordinates • trace a circular arc While it is true that a straight line is an arc with infinite radius, and rotation in place is an arc with zero radius, these special cases are the most commonly used and each deserves its own method. This allows the amount of movement to be specified by a single parameter.
The various tasks involved in navigation can usefully be classified by their complexity. 1. Execute elemental movements 2. Update the robot pose after each elemental movement, assuming an initial pose is known. In the absence of sensor data from external sources, this dead reckoning navigation and the errors of pose estimation increases with each move. 3. Improve the accuracy of the estimated pose by using sensor data and external landmarks, using algorithms such Kalman filter and extensions, and Monte Carlo Localization based on particle sets. 4. More complex tasks such as global localization, path planning, obstacle avoidance and combinations of these. I have not thought much about these, so will not discuss them. My view is that the classes and interfaces that implement navigation tasks at each level should be built on the level below. In the following, I use the language of agents that have skills of their own, and use other agents as helpers. I hope this focuses on concepts and not on the details of the API.
Level 1 Basic classes and interfaces BasicPilot interface
- Skills: travel in a straight line : forward or backward) ( I don’t think this is necessary in a Navigation package since a robot *that cannot change direction cannot it cannot navigate in a plane and is hardly worth building)
Arc Pilot interface
- Executes two of the elementary movements in a plane: straight line and arc move, forward and backward . Movement can be interrupted by a client.
Provide estimate, at any time, of distance traveled and heading change since start of the current movement.
Pilot interface Skills: Executes two of the elementary movements in a plane: straight line and arc move, forward and backward . Movement can be interrupted by a client. Provide estimate, at any time, of distance traveled and heading change since start of the current movement Uses: Motor
DifferentialPilot class (renamed from TachoPilot ) implements Pilot Skills: See Pilot Uses: two TachoMotors for differential steering;
Level 2A ----------- basic update of pose
Pose Skills: update x,y, and heading using distance traveled and heading change Calculate distance and heading to a point Uses: lejos.geom.Point, Math Note: To update this simple pose requires a navigator class that uses a gets the distance and heading change from a Pilot and calls the update method on the Pose
Level 2 ----------- automatic update of pose at the end of each elemental movement.
Movement class Skills: none –a passive data carrier of distance traveled, heading change and movement type as supplied by MovementProvider , used by a MoveListener
MovementProvider interface Skills: updates MovementListeners at the end of every elemental movement and when requested by a MovementListener Uses: Movement , MovementListener
Interface: MovementListener Skills: updates itself from data contained in a Movement which is supplied by a MovementProvider Uses: Movement , MovementProvider to supply a current Movement
UpdatingPilot interface extends Pilot implements MovementProvider Skills: See MovementProvider and Pilot Uses Movement, Motor
UpdatingDifferntialPilot class implements Updataing Pilot extends DifferentialPilot
Skills: see UpdataingPilot and DifferentialPilot
UpdateablePose extends Pose implements MoveListener Skills: See Pose and MoveListener Uses: Movement
SimpleNavigator (stripped down version of current class only has goTo(x,y) Skills: moves the robot to a desired Point. Uses: UpdatingPilot, UpdateablePose
RCNavigator extends SimpleNavigator Uses; communicator to get commands from a remote controller.
ProbabisticMovement extends Movement ***( could be called QualifiedMovement if you really want to) Skills: carries three variable Normal distribution representation of a movement;
Interface: ProbabilisticMovementProvider extends MovementProvider Skills: updates ProbabilisticMoveListeners Uses: ProbabilisticMovement
ProbabilisticPilot extends UpdatingPilot implements ProbabilisticMovementProvider Skills: see ProbabilisticMovementProvider see UpdatingPilot
ProbabilisticDifferentialPilot extends UpdatingDifferentialPilot implements ProbabilisticMovementProvider Skills: see ProbabilisticMovementProvider see UpdatingDifferentialPilot
NormalPose extends UpdateablePose Skills: uses a three variable normal distribution to represent pose. Returns x,y and heading with mean and variance.
ParticleSetPose extends UpdateablePose Skills: uses a particle set to represent a pose. Can return mean and variance of x,y and heading.
KalmanPoseProvider (probably an abstract class with sub classes for different sensor models)
- Update a NormalPose using data from external sensors;
- various sensors and motors to direct the sensors;
- Navigator (to aim the robot if sensors are fixed)
- Updating algorithm using Kalman filter
Map (to determine where to point sensors) MonteCarloPoseProvider (probably an abstract class with sub classes for different sensor models) Skills: updates a ParticleSetPose Uses: ParticleSetPose various sensors and motors to direct the sensors; Navigator (to aim the robot if sensors are fixed) Uses Monte Carlo update. Map (to determine where to point sensors)
KalmanNavigator extends SimpleNavigator Skills: moves the robot to a desired destination Point as accurately as possible Uses: ProbabilisticPilot, NormalPose, KalmanPoseProvider
MonteCarloNavigator Skills: moves the robot to a desired destination Point as accurately as possible Uses: ProbabilisticPilot, ParticleSetPose, MonteCarloPoseProvider
Higher levels of complexity - ideas gleaned form other posts as I understand them.
GlobalLocalizer Skills: uses a Map to determine the robot position with minimum initial information Uses: Navigator, ProbabilisticPilot, ParticleSetPose, MonteCarlotPoseProvider `
PathPlanner (tentative not sure of a good collaboration with Obstacle avoider) Skills: determine and follow the best route (a series of waypoints ) to get to a specified destination. Uses: Map Navigator ( goTo( next waypoint) Pose. getX(), getY() Obstacle avoider (tentative – not sure of a good relationship with Path Planner) Skills: go to specified destination despite obstacles detect and avoid obstacles. Uses: Navigator (for maneuvering around obstacle) PathPlanner (for next waypoint to avoid the obstacle ?) Map (update with representation of obstacle?)
Pose . getX(), getY()
Map Represents things in a plane, such as walls and landmarks Skills ? Distance to nearest thing (in a specified direction) from (x,y) Direction to the closest point in a wall from (x,y) Others??