CHAPTER Calinon_ISRR_2018/IDIAP Robot Learning with Task-Parameterized Generative Models Calinon, Sylvain Bicchi, A. Ed. Burgard, W. Ed. EXTERNAL https://publications.idiap.ch/attachments/papers/2017/Calinon_ISRR_2018.pdf PUBLIC https://publications.idiap.ch/index.php/publications/showcite/Calinon_ISRR_2015 Related documents Robotics Research 2 111-126 978-3-319-60916-4 2018 Springer https://doi.org/10.1007/978-3-319-60916-4_7 URL 10.1007/978-3-319-60916-4_7 doi Task-parameterized models provide a representation of movement/behavior that can adapt to a set of task parameters describing the current situation encountered by the robot, such as location of objects or landmarks in its workspace. This paper gives an overview of the task-parameterized Gaussian mixture model (TP-GMM) introduced in previous publications, and introduces a number of extensions and ongoing challenges required to move the approach toward unconstrained environments. In particular, it discusses its generalization capability and the handling of movements with a high number of degrees of freedom. It then shows that the method is not restricted to movements in task space, but that it can also be exploited to handle constraints in joint space, including priority constraints. CONF Calinon_ISRR_2015/IDIAP Robot Learning with Task-Parameterized Generative Models Calinon, Sylvain Proc. Intl Symp. on Robotics Research 2015 Task-parameterized models provide a representation of movement/behavior that can adapt to a set of task parameters describing the current situation encountered by the robot, such as location of objects or landmarks in its workspace. This paper gives an overview of the task-parameterized Gaussian mixture model (TP-GMM) introduced in previous publications, and introduces a number of extensions and ongoing challenges required to move the approach toward unconstrained environments. In particular, it discusses its generalization capability and the handling of movements with a high number of degrees of freedom. It then shows that the method is not restricted to movements in task space, but that it can also be exploited to handle constraints in joint space, including priority constraints.