Lab 05

During this lab, you should learn how to work with factorgraph-based SLAM and how to run it in the simulator.

Relevant lectures: 00_1d_mle.pdf, 00_2d_mle.pdf.

https://www.youtube.com/live/YCE1Aj0k1UA?feature=share&t=2045

The whole Tartan SLAM Series is a great study material for those who want to dive deep into how SLAM in 3D is done in state-of-the-art robotics.

What is the difference between (E)KF SLAM and Factorgraph SLAM?

aro_hw4_2024.pdf

What would the residuals and Jacobian entries look like?

• Reasonable factors:
  • Global absolute localization (GNSS, Vicon, RFID): $res_t^{gps} = ?$
    • 2-DOF, 3-DOF
  • Compass: $res_t^{compass} = ?$
  • Absolute pose priors: $res_t^{prior} = ?$
  • Interpolate marker measurement between two poses for better precision: $res_t^{mri} = ?$
  • Motion model (e.g. differential drive model): $res_t^{motion} = ?$
    • How to construct the model if $u_t$ are wheel velocities?
  • Loop closures: $res_t^{loop} = ?$
  • Velocity measurements in body frame: $res_t^{vel} = ?$
  • UWB localization (radio beacons with distance measurement): $res_t^{uwb} = ?$
  • UWB relative marker: $res_t^{uwbm} = ?$
  • Bluetooth detection (radio beacons without distance measurement): $res_t^{bt} = ?$
    • This introduces inequality constraints which are generally not very well handled.
    • You can use a robust loss to approximate the inequality.
    • Or you can pass the inequality bounds to the bounds parameter of least_squares.
  • Marker as LED in camera (cannot tell its distance): $res_t^{led} = ?$
• Silly factors (RUR Challenge). Figure out a sensor that could use them.
  • Slow light propagation: Marker detections are delayed.
  • Gravity field changing in space (can you try to map it?).
  • Time “speed” distortion changes with position (+ sensor that measures sin(distortion)). (Darius Diebold)
  • Gravitational field that changes rapidly with position on surface (+ sensor that measures sin(||x||)). (Laura G. Hernandez)
  • Wavy gravity field (+ sensor that measures sin(x*x)). (Marc Simon)
  • Mass of drilling robot increases exponentially with depth (+ sensor that measures x*exp(-x)). (Jan Kubis)
  • Space-dependent time distortion (+ sensor that measures x+sin(x+distortion)). (Stanislav Kunst)
  • Gravity depends on sin(x) (+ sensor that observes a landmark and measures sign of gravity). (Michal Meciar)
  • Pose updates with x + sin(x) + u (+ sensor that measures log(1 + x*x)). (Mark Horpynych)
  • Gravity decreases with r^(1.5), (+ sensor that localizes known black holes). (Stefan Kando)
  • Light propagates according to c * log(t) (+ sensor that measures exp(||x||) and angle to markers). (Votjech Klouda)
  • Gravity field is influenced by cosine of position (+ sensor that measures sin(omega * x)). (Matej Mihulka)
  • Strong magnetic field causing observable Faraday rotation (+ sensor that measures 1/x). (Tomas Ort)
  • Magic crystals slowing down time (+ sensor that measures 1/||x-m||). (Valeriia Timonova)
  • Gravity as a sine wave (+ time-delayed sensor that measures sin(x-m)). (Shaawaiz Haider)
  • Teleportation world (+ sensor measuring sine of position). (Theodoros Kokkoris)
  • Time-warping on ash-filled orbit (+ sensor that measures sin(x)). (Muhammad Moeed Zeb)
  • Non-Newtonian fog (+ sensor that measures exp(|x-m|^2)). (Adam Cetlovsky)
  • Time-varying gravity (+ drunken shepherd sensor that measures gravity). (Jozef Gal)
  • World with waving string interaction strength, materials transmuting all the time (+ sensor that measures ln(1 + x)). (Krystof Svejda)
  • Gas giant world (+ sensor measuring (x-m)^2). (Vit Zoubek)
  • Objects with same phases attracted (+ sensor that measures sin(r+phi)). (Patrik Martisko)
  • Friction is inversely proportional to sound level (+ sensor that measures acoustic slip). (Kyaw Htet Lin)

Read and try to understand the assignment of the homework HW4.