Perceive and locate features of importance

Introduction to autonomous navigation techniques for intelligent vehicles

Figure 1 - Reference marks RE (ECEF) in R0 (ENU). Note the spherical approximation of the Earth in this model, which introduces no error when changing reference points

Figure 4 - Marker RM linked to the vehicle and marker R0 tangent to earth

Figure 7 - The map-matching problem. The aim is to determine the point on the polyline closest to the origin of the vehicle&#039;s reference frame.

Figure 10 - Construction of the navigable space for the vehicle. The rolling space is in dotted black, collision zones are in red and the resulting navigable space is in blue.

Figure 11 - Dirichlet decomposition (Voronoi diagram). The limits of the navigable space are indicated by the black zones.

Figure 12 - Visibility graph. Nodes (in orange) are connected by branches (in black) only if they are contained in the navigable space whose borders are in black.

Figure 14 - Path (in blue) obtained after smoothing the result obtained by the Dijkstra algorithm in the visibility graph

Figure 17 - Lateral control along a horizontal segment. The distance L between the front and rear wheels is not shown.