DEPRECATED - PLEASE USE OWL INSTEAD

The repository contains rapid prototype implementations of motion planners and their variants, version 0.1.0.

The code includes experimental features that in their current form should not make their way to a live robot. We have extracted and refined the approved methods to the repository owl.

List of algorithms: GLC, RRT*

The references are

• A Generalized Label Correcting Method for Optimal Kinodynamic Motion Planning by Brian Paden and Emilio Frazzoli, arXiv:1607.06966
• Sampling-based algorithms for optimal motion planning by Sertac Karaman and Emilio Frazzoli, IJRR11

The following integrators are available:

• Euler, Midpoint
• Runge-Kutta 4th order, and 5th order
• exact integrator for the group SE2

The owly repository implements visualizations in 2D as showcased below. See also a video.

The separate repository owly3d implements animations and visualizations in 3D.

Examples

GLC

Rice2: 4-dimensional state space + time

---

SE2: 3-dimensional state space

Car	Two-wheel drive (with Lidar simulator)

---

Tracking of potential locations of pedestrians and vehicles

---

Pendulum Swing Up

---

Rice1

---

R^2

---

Lotka-Volterra: predator and prey, control by decay rate of predators

---

surface flow in river delta

against the direction of current

current reversed

AnyTime GLC

R^2

RRT*

R^2

Contributors

Jan Hakenberg, Jonas Londschien, Yannik Nager