Learning from demonstration and adaptation of biped locomotion

By Jun Nakanishi, Jun Morimoto, Gen Endo, Gordon Cheng, Stefan Schaal, Mitsuo Kawato. 2004

487 citations recorded by the source corpus.

Subfields: Biomedical Engineering, Cognitive Neuroscience, 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 Biomedical Engineering

In Cognitive Neuroscience

In Control and Systems Engineering

Citations in the corpus

Overcoming Exploration in Reinforcement Learning with Demonstrations (2017, background): “BC has seen success in autonomous driving [5], [6], quadcopter navigation [7], locomotion [8], [9].”
Learning perceptual coupling for motor primitives (2008, background): “Resulting successes have shown that it is possible to rapidly learn motor primitives for complex behaviors such as tennis swings [1], [2], T-ball batting [5], drumming [6], biped locomotion [3], [7] and even in tasks with potential industrial application [8].”
Reinforcement learning of motor skills with policy gradients (2008, methodology): “Adjustment of the primitives using sensory input can be incorporated by modifying the internal state z of the system as shown in the context of drumming (Pongas, Billard, & Schaal, 2005) and biped locomotion (Nakanishi et al., 2004; Schaal, Peters, Nakanishi, & Ijspeert, 2004).”
SIMBICON: simple biped locomotion control (2007, methodology): “It has been applied in a number of ways to the control of walking[Tedrake et al. 2004; Nakanishi et al. 2003; Morimoto et al. 2004; Smith 1998].”
Learning Movement Primitives (2005, methodology): “As a last evaluation of DMPs, we applied the phase oscillator DMP to simulated biped locomotion [ 29 ] of a planar biped (Figure 4). Motion capture data from human locomotion was employed to learn an initial trajectory pattern, which, after some modest tuning of the speed and amplitude parameters of”
A Multi-Facet Survey on Memetic Computation (2011, background): “Some other memetic representations include the locally weighted regression [186] and the self-model proposed in [187] for facilitating the imitation of memes in the form of complex biped locomotion movement from human experts, thus enabling agents to diagnose and recover from unanticipated situation”
Programmable central pattern generators: an application to biped locomotion control (2006, background): “Importance of phase resetting for biped locomotion was already discussed in [ 11 ] where they showed that it creates entrainment of the controller with the body dynamics of the robot.”
Incremental learning of full body motion primitives and their sequencing through human motion observation (2012, background): “2006), phase oscillators for rhythmic motions (Nakanishi et al. 2004), dynamical models (Ijspeert et al.”
Learning motor primitives for robotics (2009, background): “Previous applications include a variety of different basic motor skills such as tennis swings [2], T-ball batting [6], drumming [7], planar biped walking [3], [8], constrained reaching tasks [9] and even in tasks with potential industrial application [10].”
Movement templates for learning of hitting and batting (2010, background): “Resulting successes have shown that it is possible to rapidly learn motor primitives for complex behaviors such as tennis swings [1] with only a ﬁnal target, constrained reaching [4], drumming [5], biped locomotion [2], [6] and even in tasks with potential industrial application [7].”
Learning grounded finite-state representations from unstructured demonstrations (2015, background): “LfD has had many successes in teaching robots tennis swings (Schaal, 2003), walking gaits (Nakanishi et al., 2004), pick-and-place tasks (Mühlig et al.”
Imitation Learning (2017, background): “…are batting tasks [Kober and Peters 2009c; Ijspeert et al. 2002b], where the agent is required to execute a motion trajectory in order to pass by a point and hit a target, and walking [Nakanishi et al. 2004], where the agent needs to produce a trajectory that results in smooth stable motion.”
Machine learning of motor skills for robotics (2007, methodology): “Supervised learning on gaits for biped locomotion has already been applied (Schaal, Peters, Nakanishi, & Ijspeert, 2004; Nakanishi et al., 2004) and we intend to use our reinforcement learning techniques on improving the resulting gaits.”
A Dynamical System-based Approach to Modeling Stable Robot Control Policies via Imitation Learning (2012, methodology): “Later approaches highlight the use of DMP for different robotics applications such as: walking (Nakanishi et al., 2004; Schaal, Peters, et al., 2003), drumming (Ude et al.”
Kernelized movement primitives (2017, unknown): “In many robotic tasks, such as biped locomotion (Nakanishi 2004) and striking movements (Huang et al. 2016), time plays a critical role when generating movement trajectories for a robot.”
Incremental Learning, Clustering and Hierarchy Formation of Whole Body Motion Patterns using Adaptive Hidden Markov Chains (2008, methodology): “Several different techniques have been applied to the modeling the motion primitives, including dynamical models (Nakanishi et al. 2004), neural networks (Ogata et al. 2005), and stochastic models (Inamura et al. 2004 Billard et al. 2006 Taylor et al. 2006).”