quartz/content/notes/10-orientation-sensors.md

title	tags
10-orientation-sensors	lecture info305

[!INFO] Location sensors dont know what direction you are looking

Accelerometers

• Measures proper acceleration (acceleration it experiences relative to freefall. gravity), felt by people or objects
• Units: m/s2 or g
• Most smartphone accelerometers trade large value range for high precision
  • iPhone range: ±2g, precision 0.018g

[!INFO] measure accelerationi with respect to gravity measure gravity along three axis tradeoff large range with high precision

Accelerometers

• Acceleration is measured on 3 axes
• Orientation of sensor (and coordinate system) varies among different device

Accelerometers

• Miniaturisation using a MEMS (Microelectromechanical systems)
• Measuring flowing current over an differential capacitor indicates the acceleration

[!INFO] dont need to fully understand how MEMS work

Accelerometers

• Advantage:
  • Fast update rate
  • Relatively accurate
• Disadvantage:
  • Can’t easily identify certain kind of acceleration (small value range)

[!INFO] only give you direction to the center of the earth (from gravity)

Compass / Magnetometer

• Measures the strength of earth’s magnetic field
• Strength is expressed in tesla [T]
• iPhone 4 magnetometer range: ±2mT
• Pro tip: prolonged exposure to a fridge magnet decalibrates your phones’ magnetometer for at least a week ;-)

Compass / Magnetometer

• Advantage:
  • Absolute orientation measurement
• Disadvantage:
  • Usually slow update
  • Sensitive to errors
    • Local discontinuities in magnetic field
    • Ferromagnetic materials
    • Power sources

[!INFO] doesn't directly measure north measure magentic fielf of earth and aligns with magenetic north earth field is very weak so we need sensitive magnetometer electronic from phone affect sensor but the error can be calibrated out balance between what is cheap and what is good enough only one degree on freedom (bearing to north)

Gyroscope

• Detects the current orientation of the device, or changes in the orientation
• Precisely: orientation can be computed from the angular rate that is detected by the gyroscope, expressed in rad/s [deg/s] on 3 axis.
• Traditional gyroscopes use the e!ect of angular momentum

[!INFO] car, phone, game controller, etc measure orientation measure change of orientation using angular rate use the effect of angular momentum. when you spin something fast, it want to keep that rotation axis

• MEMS (microelectromechanical system) gyro
• Used in most smartphones or mobile/embedded devices
• Use the displacement of vibrating proof mass to compute orientation (Coriolis effect)

[!INFO] works similar to spinning gyro but uses vibrations instead (coriolis effect)

Problem:

• The gyroscope gives us angular rate with a unit of rad/s [deg/s]
• We can find the angular position at any given moment t with the following equation (assuming t=0 theta=0)
• We cannot take a perfectly continuous integral -> take the sum of a finite number of samples taken at a constant interval Ts
• Gyroscope data changes faster than the sampling frequency
  • We will not detect it, and the integral approximation will be incorrect
  • This error is called drift/bias as it increases in time, no return to 0o

[!INFO] dont need to fully understand the math

Recap

• Several improvements to traditional GPS

• AGPS for improved startup time and improved localisation using WIFI

• DGPS for improved localisation using reference stations with known error

• RTK GPS for improve localisation using DGPS and phase analysis

• Accelerometers for measuring gravity along multiple axis (typically 3)

• Usually implemented with MEMS

• Typically limited in value range (in mobile devices)