www.wiki.ardumower.de - Benutzerbeiträge [de]

Lawn sensor

2015-01-06T12:33:17Z

Frank: While technically "capacity" is a correct description of the physical effect, "capacitance" is much more common in English usage.

== Lawn Sensor ==

[[File:capacity.png|caption]]

A lawn sensor should detect if the robot is still on lawn or not. In other words, it should find out if the media directly below the robot is still lawn or something else (air etc.). Therefore, a capacitance sensor can be used as the media below the robot has influence on the capacitance. However, there is no difference in capacitance between lawn and flowers etc. - so you should have a distinct area around your lawn that is different to lawn (e.g. bark mulch).

== Principle of operation ==

A lawn sensor could be a copper plate in a closed plastic tube. To find out if the plate has contact with lawn, the capacitance of this sensor is measured. If the robot is on lawn (lawn is 'dielectric'), the measured capacitance increases slightly (by a few pF).

<gallery>
File: l50_capmeter.jpg

File: lawn_sensor.jpg

File: capsense.gif
</gallery>

== Wiring: ==

'''''Front lawn sensor:'''''

Arduino Mega Pin 40 Receive Front -------------|

Arduino Mega Pin 41 Send Front --- 2 MOhm --+---| Lawn

'''''Back lawn sensor:'''''

Arduino Mega Pin 42 Receive Back -------------|

Arduino Mega Pin 43 Send Back --- 2 MOhm --+---| Lawn

== Detailed operation ==

Initially, the copper plate is discharged by pulling the send pin to low.
Then the send pin is set to 'high' - via a resistor (2 Megaohm) the copper plate is charging slowly.
By constantly reading input receiver pin, the Arduino measures the time until it becomes 'high' (capacitance measurement).
Because microcontrollers are fast and precise in time measurement, you can measure very low capacities with this.

GPS

2014-12-30T00:05:50Z

Frank: Minor English usage errors corrected...

=Abstract=
<gallery>
File: ublox_neo_6m.jpg

File: GY-NEO6MV2_schematics.jpg
</gallery>

With the help of a GPS receiver (e.g. GY-NEO6MV2, ublox 6m), the long-term position can be calculated. Therefore, the GPS position values will be averaged.

Currently, GPS is used
* to receive current date and time

= Wiring: =
GPS TX (3.3V compatible) -- Arduino RX
GPS RX (3.3V compatible) -- Arduino TX
GPS VCC -- Arduino VCC 5V
GPS GND -- Arduino GND

= Visualization of GPS data =
GPS data (course, speed, position etc.) can be plotted by pfodApp:

<gallery>
File: gpsplot.jpg
File: gpsplotting.jpg
File: gpstracking.jpg
</gallery>

= Fusion of sensor values =
# Calculation of short-time position by Odometry and Compass
# Calculation of long-term position via GPS

<gallery>
File: ardumower_odometry.png
</gallery>

A Kalman filter can be used to fuse all sensor values.

= Further links =

Ardumower software design

2014-12-29T23:52:45Z

Frank: /* Finite state machine */

This page describes some internals of the Ardumower software architecture.

=Finite state machine=
Main component of the software is a so called 'finite state machine', that means there exists a set of states ("OFF", "FORWARD", "ROLL", etc.) that the robot can be in. Depending on events (sensor is triggering etc.), the robot will enter a new state.

State diagram

<gallery>
File: Ardumower_statemachine.png
</gallery>

Implemented updates:
*If obstacle during reverse => roll
*If obstacle during roll => forward

=Further links=