Is potential field or navigation potential function suitable for obstacle avoidance for optimal path planning of mobile robot using optimal control approach.
Vijay I am not sure if I follow, kindly elaborate. Thank you Karnika and Soumic, will check the paper out.
I'm not sure I understand what you mean by "optimal path planning of mobile robot using optimal control approach," but in order to ensure optimal path planning, you would need to use an additional path planning algorithm that guarantees optimality, such as A* to find the optimal path, and then pick waypoints along that path to use as goal points for a potential field approach to motion planning. Karnika is right that you need global information of the environment in this case.
You will still have the local minima problem with the potential field to worry about, but if you pick good waypoints it can help you avoid the local minima that can cause problems.
Alternatively, you can use a potential field on heading and it's possible to get near-optimal paths to emerge from these control laws. You still have a number of practical issues to address, but it can work reasonably well. These papers present a few examples of this
http://people.csail.mit.edu/lpk/papers/fajen-warren-temizer-lpk.pdf
http://www.cog.brown.edu/research/ven_lab/Publications/Huang_Robot_RAS06.pdf
http://www.jan-wiener.net/publications/Gerstmayr_Mallot_Wiener-2008.pdf
Feel free to take a look at this paper:
Conference Paper Mobile Robot Navigation System Based on Probabilistic Road M...
- Badges
- Science topic
- Similar topics
- Higher Education
- Universities