11. Grid-based FastSLAM Techniques

Grid-based FastSLAM Techniques

To adapt FastSLAM to grid mapping, we need three different techniques:

  1. Sampling Motion-<span class="mathquill ud-math">p(x_{t} | x_{t-1}^{[k]} , u_{t})</span>: Estimates the current pose given the k-th particle previous pose and the current controls u.
  2. Map Estimation-<span class="mathquill ud-math">p(m_{t} | z_{t}, x_{t}^{[k]} , m_{t-1}^{[k]})</span>: Estimates the current map given the current measurements, the current k-th particle pose, and the previous k-th particle map
  3. Importance Weight-<span class="mathquill ud-math">p(z_{t} | x_{t}^{[k]} , m^{[k]})</span>: Estimates the current likelihood of the measurement given the current k-th particle pose and the current k-th particle map.

How is each of these techniques solved under the Grid-based FastSLAM algorithm?
SOLUTION:
  • **Sampling motion** technique can be solved with **MCL**.
  • **Importance weight** technique can be solved with **MCL**.
  • **Map estimation** technique can be solved with **Occupancy Grid Mapping**.