Rich periodic motor skills on humanoid robots: Riding the pedal racer

Gams, Andrej; Kieboom, Jesse van den; Vespignani, Massimo; Guyot, Luc; Ude, Aleš; Ijspeert, Auke Jan

doi:10.1109/icra.2014.6907181

Cited by 8 publications

(6 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…They also evaluated their work on a whiteboard cleaning task (Kormushev et al 2011c). A similar task is surface wiping which is investigated by Urbanek et al (2004); Gams et al (2014). Both works represent the wiping skill as a periodic movement.…”

Section: Spatially Constrained Behavior (B)mentioning

confidence: 99%

“…In this case, rhythmic motions are advantageous, as the complete surface can be wiped easily by executing the motion several times while shifting only the center point. The work from Gams et al (2014) also uses force feedback to maintain contact with the surface. Besides the aforementioned household tasks, there are also industrial operations that require constant environmental contact.…”

Section: Spatially Constrained Behavior (B)mentioning

confidence: 99%

See 1 more Smart Citation

A Survey of Behavior Learning Applications in Robotics -- State of the Art and Perspectives

Fabisch,

Petzoldt,

Otto

et al. 2019

Preprint

View full text Add to dashboard Cite

Recent success of machine learning in many domains has been overwhelming, which often leads to false expectations regarding the capabilities of behavior learning in robotics. In this survey, we analyze the current state of machine learning for robotic behaviors. We will give a broad overview of behaviors that have been learned and used on real robots. Our focus is on kinematically or sensorially complex robots. That includes humanoid robots or parts of humanoid robots, for example, legged robots or robotic arms. We will classify presented behaviors according to various categories and we will draw conclusions about what can be learned and what should be learned. Furthermore, we will give an outlook on problems that are challenging today but might be solved by machine learning in the future and argue that classical robotics and other approaches from artificial intelligence should be integrated more with machine learning to form complete, autonomous systems.

show abstract

Section: Spatially Constrained Behavior (B)mentioning

confidence: 99%

Section: Spatially Constrained Behavior (B)mentioning

confidence: 99%

A Survey of Behavior Learning Applications in Robotics -- State of the Art and Perspectives

Fabisch,

Petzoldt,

Otto

et al. 2019

Preprint

View full text Add to dashboard Cite

show abstract

“…Instead of making repeated runs of a desired finite time trajectory, RC aims to perfectly execute a periodic command or to execute a periodic command in the presence of a periodic disturbance. Repetitive control was, for example, employed on a task of riding a pedal racer which demands balancing of a humanoid robot [24]; see also Fig. 14. In this work, periodic dynamical movement primitives are combined with RC according to force feedback.…”

Section: Iterative Learning Control (Ilc)mentioning

confidence: 99%

“…Although the feasibility of balancing on a pedal racer was demonstrated with the COMAN humanoid robot in [24], there are additional challenges in bipedal walking. In the former case, the robot's feet keep the contact with the pedal racer during riding, and the feet trajectory is limited to a motion primitive constrained by the kinematics of the pedal.…”

Section: Iterative Learning Control (Ilc)mentioning

confidence: 99%

Learning Control

Calinon

Lee

2018

Humanoid Robotics: A Reference

View full text Add to dashboard Cite

This chapter presents an overview of learning approaches for the acquisition of controllers and movement skills in humanoid robots. The term learning control refers to the process of acquiring a control strategy to achieve a task. While the definition is in some cases restrained to trialand-error learning, we present here learning control in a broader perspective, with a focus on the representation of skills to be acquired, and on the different learning strategies that can contribute to the acquisition of robust and adaptive controllers for humanoids.

show abstract

“…9(b)]. We used periodic DMPs to represent the movement [49] and apply a Repetitive Control (RC) algorithm [50], [51]. The RC algorithm iteratively adapts the current estimate of the stirring behavior to achieve the desired outcome as defined by the desired contact force.…”

Section: A Bootstrapping Motion -Measuring Learning Speedmentioning

confidence: 99%

Structural Bootstrapping—A Novel, Generative Mechanism for Faster and More Efficient Acquisition of Action-Knowledge

Wörgötter¹,

Geib

Tamošiūnaitė

et al. 2015

IEEE Trans. Auton. Mental Dev.

Self Cite

View full text Add to dashboard Cite

Humans, but also robots, learn to improve their behavior. Without existing knowledge, learning either needs to be explorative and, thus, slow or-to be more efficient-it needs to rely on supervision, which may not always be available. However, once some knowledge base exists an agent can make use of it to improve learning efficiency and speed. This happens for our children at the age of around three when they very quickly begin to assimilate new information by making guided guesses how this fits to their prior knowledge. This is a very efficient generative learning mechanism in the sense that the existing knowledge is generalized into as-yet unexplored, novel domains. So far generative learning has not been employed for robots and robot learning remains to be a slow and tedious process. The goal of the current study is to devise for the first time a general framework for a generative process that will improve learning and which can be applied at all different levels of the robot's cognitive architecture. To this end, we introduce the concept of structural bootstrapping-borrowed and modified from child language acquisition-to define a probabilistic process that uses existing knowledge together with new observations to supplement our robot's data-base with missing information about planning-, object-, as well as, action-relevant entities. In a kitchen scenario, we use the example of making batter by pouring and mixing two components and show that the agent can efficiently acquire new knowledge about planning operators, objects as well as required motor pattern for stirring by structural bootstrapping. Some benchmarks are shown, too, that demonstrate how structural bootstrapping improves performance.Index Terms-Fast learning, generative model, knowledge acquisition. 1943-0604

show abstract

Rich periodic motor skills on humanoid robots: Riding the pedal racer

Cited by 8 publications

References 25 publications

A Survey of Behavior Learning Applications in Robotics -- State of the Art and Perspectives

A Survey of Behavior Learning Applications in Robotics -- State of the Art and Perspectives

Learning Control

Structural Bootstrapping—A Novel, Generative Mechanism for Faster and More Efficient Acquisition of Action-Knowledge

Contact Info

Product

Resources

About