Interface Capture

All Known Implementing Classes:

public interface Capture
extends SimpleN150, VisualObjectSensor, GripperActuator, KinSensor, Transceiver

Provides an abstract interface to the hardware of a foraging Nomad 150 robot. Introduction
If you write a control system using this interface to the hardware, you can test it in simulation and on mobile robots.

Capture robots are multiforaging robots with long-range vision based on the Nomad 150 platform. "Multi Foraging" means the robot can sense several different classes of objects to collect, along with different bins in which to to deposit them. A SimpleN150 robot has sonar range finders, bumper switches, and odometry and steering, translation and turret motors. These capabilities are extended on a MultiForageN150 through the addition of color vision and a gripper.

Vision sensing
Vision hardware provides six "channels" that each track a different type of object. Each call to one of the vision routines requires a reference to which channel is being accessed. Bins and objects to collect are sensed by the same vision hardware.

Frames of reference
We use a standard cartesian coordinate system in meters and radians. Pretend you are looking down on the robot: +x goes out to your right (East), +y goes up (North). When the robot is initialized, it is facing the +x direction. Headings are given in radians, with East=0, North=PI/2 and so on CCW around to 2*PI. Some methods return "egocentric" vectors. An egocentric vector is given relative to the center of the robot in the same heading reference frame as global coordinates. An object one meter east of the robot is at (1,0) egocentrically.

This class is extended by a simulation class (CaptureSim). The subclasse handles details of interaction with the simulated world.

Many of the sensor and motor command methods (e.g. get* and set*) require a timestamp as a parameter. This is to help reduce the amount of I/O to the physical robot. If the timestamp is less than or equal to the value sent on the last call to one of these methods, old data is returned. If the timestamp is -1 or greater than the last timestamp, the robot is queried, and new data is returned. The idea is that during each control cycle the higher level software will increment the timestamp and use it for all calls to these methods.

Copyright (c)1997, 1998 Tucker Balch

See Also:

Field Summary
static double GRIPPER_POSITION
static int VISION_FOV_DEG
static double VISION_FOV_RAD
static double VISION_RANGE
Fields inherited from interface
Methods inherited from interface
getTurretHeading, resetTurretHeading, setTurretHeading
Methods inherited from interface
getVisualObjects, setVisionNoise
Methods inherited from interface
getObjectInGripper, setGripperFingers, setGripperHeight
Methods inherited from interface
getOpponents, getPlayerNumber, getTeammates, setKinMaxRange
Methods inherited from interface
broadcast, connected, getReceiveChannel, multicast, quit, setCommunicationMaxRange, unicast
Methods inherited from interface
getBackgroundColor, getDictionary, getForegroundColor, getID, getID, getObstacles, getPosition, getSteerHeading, getTime, resetPosition, resetSteerHeading, setBaseSpeed, setDictionary, setDisplayString, setID, setObstacleMaxRange, setSpeed, setSteerHeading

Field Detail


public static final double VISION_RANGE


public static final int VISION_FOV_DEG


public static final double VISION_FOV_RAD


public static final double GRIPPER_CAPTURE_RADIUS


public static final double GRIPPER_POSITION