Authors:
Giuseppe Zuffanti
1
and
Costantino Scozzafava
2
Affiliations:
1
Institute for Scientific Interchange Foundation and University of Turin, Italy
;
2
Erxa S.r.l., Italy
Keyword(s):
Path planning, Motion planning, Roadmap graph, Potential field, Look-ahead, Obstacle avoidance, Attraction point, Dynamic environment.
Abstract:
In this work we focus on the problem of 2D path planning and motion planning for a mobile robot. The goodness of a path planning and motion planning methods can be evaluated primarily through the parameters such as safety of generated trajectory speeds and robustness against dynamic changes of the environment. In all applications that use mobile robots, the motion planning problem is of crucial importance. According to different applications one can put emphasis on the previous characteristics listed, potentially to the detriment of the others. Typically, methods that generate high trajectory speeds for the navigating robot appears to be less robust and safe. The importance that the path planning and the motion planning problem play in practical applications is also demonstrated by the several techniques developed to tackle them. We propose a complete framework which creates a clear shortest path for a robot in an environment with static obstacles and generates, in real time, the acc
urate trajectory taking into account the presence of dynamics obstacles. Application of the framework to the case of a differential drive robot in a dynamic indoor environment is shown and the proposed approach has been tested both with simulation and real data.
(More)