On-line regression algorithms for learning mechanical models of robots: A survey

By Olivier Sigaud, Camille Salaün, Vincent Padois. 2011

150 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Control and Systems Engineering

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Control and Systems Engineering

Citations in the corpus

Active learning of inverse models with intrinsically motivated goal exploration in robots (2013, neutral): “Yet, these regression mechanisms are efficient only if the quality and quantity of data is high enough, which is not the case when using unconstrained exploration such as random exploration.”
On-line policy learning and adaptation for real-time personalization of an artificial pancreas (2015, methodology): “Additionally, sparsiﬁcation techniques have been combined with regression techniques (Sigaud, Salaün, & Padois, 2011) in order to limit the computational complexity.”
A survey of deep learning techniques for autonomous driving (2019, background): “, 2015), (Ostafew, 2016), or Support Vector Regression (Sigaud et al., 2011).”
A Dynamical System-based Approach to Modeling Stable Robot Control Policies via Imitation Learning (2012, background): “, 2008), learning robots model (Jordan & Rumelhart, 1992; Sigaud et al., 2011; Peters & Schaal, 2008; Vijayakumar et al., 2005; Nguyen-Tuong et al., 2008; Butz et al., 2007; Jaeger et al., 2007), recognition of gestures (Ardizzone et al.”
A Bio-Inspired Incremental Learning Architecture for Applied Perceptual Problems (2016, methodology): “To perform true incremental learning in the aforementioned sense, most approaches perform a local partitioning of the input space and train a separate classification/regression model for each partition [5, 6, 7, 8, 9].”
Robust Constrained Learning-based NMPC enabling reliable mobile robot path tracking (2016, background): “Learned models are commonly modeled as GPs, Locally Weighted Projection Regression, or Support Vector Regression (Sigaud et al., 2011).”
Incremental learning algorithms and applications (2016, background): “Many modern incremental learers rely on a local partitioning of the input space, and a separate classification/regression model for each partition [160, 121, 148, 18, 21].”
Intrinsically Motivated Learning of Real-World Sensorimotor Skills with Developmental Constraints (2013, background): “As explained below, this raises the well known problem of the curse-of-dimensionality(Bishop, 1995), which needs to be addressed even for single specific skill learning involving the learning of forward and inverse models given a control space and a task space (Nguyen-Tuong and Peters, 2011; Sigaud ”
Incremental learning with self-organizing maps (2017, methodology): “To perform incremental learning as defined earlier in this section, most modern approaches perform an explicit local partitioning of the input space and train a separate classification/regression model for each partition [24], [25], [26], [27], [28].”
Value and reward based learning in neurorobots (2013, methodology): “Many methods exist which can estimate the system dynamics (Nguyen-Tuong and Peters, 2011; Sigaud et al., 2011), we adopt a simple least squares method to estimate a(x) and B(x) with basis functions.”
Cautious Model Predictive Control Using Gaussian Process Regression (2017, methodology): “Learning methods for automatically identifying dynamical models from data have gained significant attention in the past years [1] and more recently also for robotic applications [2], [3].”