To access the full text documents, please follow this link: http://hdl.handle.net/2117/177519

Title:	Anticipative kinodynamic planning: multi-objective robot navigation in urban and dynamic environments
Author:	Ferrer, Gonzalo; Sanfeliu Cortés, Alberto
Other authors:	Universitat Politècnica de Catalunya. Departament d'Enginyeria de Sistemes, Automàtica i Informàtica Industrial; Universitat Politècnica de Catalunya. VIS - Visió Artificial i Sistemes Intel.ligents
Abstract:	“This is a post-peer-review, pre-copyedit version of an article published in Autonomous robots. The final authenticated version is available online at: http://dx.doi.org/10.1007/s10514-018-9806-6”.
Abstract:	This paper presents the anticipative kinodynamic planning (AKP) approach for robot navigation in urban environments, while satisfying both dynamic and nonholonomic constraints. Our main motivation is to minimize the impact that the robot is doing to the environment, i.e. other pedestrians, while successfully achieving a navigation goal. To this end, we require a better understanding of the environment, and thus, we propose to integrate seamlessly a human motion prediction algorithm into the planning algorithm. In addition, we are able to anticipate for each of the robot’s calculated paths or actions the corresponding people’s future trajectories, which is essential to reduce the impact to nearby pedestrians. Multi-objective cost functions are proposed and we describe a well-posed procedure to build joint cost functions. Plenty of simulations and real experiments have been carried out to demonstrate the success of the AKP, compared to other navigation approaches
Abstract:	Peer Reviewed
Subject(s):	-Àrees temàtiques de la UPC::Informàtica::Robòtica -Mobile robots -Automatic control -Robots -- Programming -Robot navigation -Motion prediction -Dynamic environments -Robots mòbils -Control automàtic -Robots -- Programació
Rights:	Attribution-NonCommercial-NoDerivs 3.0 Spain http://creativecommons.org/licenses/by-nc-nd/3.0/es/
Document type:	Article - Submitted version Article
Share:

Show full item record