Autonomous Learning of Ball Trapping in the Four-Legged Robot League

Kobayashi, Hayato; Osaki, Tsugutoyo; Williams, Eric; Ishino, Akira; Shinohara, Ayumi

doi:10.1007/978-3-540-74024-7_8

Cited by 6 publications

(9 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It is a combination of the episodic SMDP Sarsa(λ) with the linear tile-coding function approximation (also known as CMAC). This is one of the most popular reinforcement learning algorithms, as seen by its use in Kobayashi et al [8]. In our experiments, we define the period from the starting to the ending of a penalty kick as one episode.…”

Section: Learning Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Development of an Augmented Environment and Autonomous Learning for Quadruped Robots

Kobayashi

Osaki

Okuyama

et al. 2009

RoboCup 2008: Robot Soccer World Cup XII

Self Cite

View full text Add to dashboard Cite

Abstract. This paper describes an interactive experimental environment for autonomous soccer robots, which is a soccer field augmented by utilizing camera input and projector output. This environment, in a sense, plays an intermediate role between simulated environments and real environments. We can simulate some parts of real environments, e.g., real objects such as robots or a ball, and reflect simulated data into the real environments, e.g., to visualize the positions on the field, so as to create a situation that allows easy debugging of robot programs. As an application in the augmented environment, we address the learning of goalie strategies on real quadruped robots in penalty kicks. Our robots learn and acquire sophisticated strategies in a fully simulated environment, and then they autonomously adapt to real environments in the augmented environment.

show abstract

Section: Learning Methodsmentioning

confidence: 99%

“…The goalie strategies involve the skills to save an incoming ball kicked by an opponent player, which is critical to not lose the game. The learning can be also regarded as the two dimensional extension of the study of Kobayashi et al [8]. They studied the autonomous learning to trap a moving ball by utilizing reinforcement learning.…”

Section: Introductionmentioning

confidence: 99%

Development of an Augmented Environment and Autonomous Learning for Quadruped Robots

Kobayashi

Osaki

Okuyama

et al. 2009

RoboCup 2008: Robot Soccer World Cup XII

Self Cite

View full text Add to dashboard Cite

show abstract

“…The final ball grasping behavior shows significant improvements over hand-coded equivalents. Similarly, Kobayashi et al [12] learn how to trap the ball using Reinforcement Learning [18]. Trapping, as opposed to grasping, is the process of intercepting and capturing a moving ball rather than a stationary one.…”

Section: Background and Related Workmentioning

confidence: 99%

“…In the last few years however, machine learning has optimized motion sequences associated with the skills of walking quickly [10,14,13], receiving passes [12], and capturing the ball [7]. The use of machine learning to optimize a skill has the benefits of removing human bias from the optimization process and, in many cases, reducing the amount of human labor required to create a top notch motion.…”

Section: Introductionmentioning

confidence: 99%

Learning Powerful Kicks on the Aibo ERS-7: The Quest for a Striker

Hausknecht

Stone

2011

RoboCup 2010: Robot Soccer World Cup XIV

View full text Add to dashboard Cite

Coordinating complex motion sequences remains a challenging task for robotics. Machine Learning has aided this process, successfully improving motion sequences such as walking and grasping. However, to the best of our knowledge, outside of simulation, learning has never been applied to the task of kicking the ball. We apply machine learning methods to optimize kick power entirely on a real robot. The resulting learned kick is significantly more powerful than the most powerful handcoded kick of one of the most successful RoboCup four-legged league teams, and is learned in a principled manner which requires very little engineering of the parameter space. Finally, model inversion is applied to the problem of creating a parameterized kick capable of kicking the ball a specified distance.

show abstract

“…The goalie strategies involve the skills to save an incoming ball kicked by an opponent player, which is critical to not lose the game. The learning can be also regarded as the two dimensional extension of the study of Kobayashi et al [14]. They studied the autonomous learning to trap a moving ball by utilizing reinforcement learning.…”

Section: Introductionmentioning

confidence: 99%

Development of an Interactive Augmented Environment and Its Application to Autonomous Learning for Quadruped Robots

Kobayashi

Osaki

Okuyama

et al. 2009

IEICE Trans. Inf. & Syst.

Self Cite

View full text Add to dashboard Cite

SUMMARYThis paper describes an interactive experimental environment for autonomous soccer robots, which is a soccer field augmented by utilizing camera input and projector output. This environment, in a sense, plays an intermediate role between simulated environments and real environments. We can simulate some parts of real environments, e.g., real objects such as robots or a ball, and reflect simulated data into the real environments, e.g., to visualize the positions on the field, so as to create a situation that allows easy debugging of robot programs. The significant point compared with analogous work is that virtual objects are touchable in this system owing to projectors. We also show the portable version of our system that does not require ceiling cameras. As an application in the augmented environment, we address the learning of goalie strategies on real quadruped robots in penalty kicks. We make our robots utilize virtual balls in order to perform only quadruped locomotion in real environments, which is quite difficult to simulate accurately. Our robots autonomously learn and acquire more beneficial strategies without human intervention in our augmented environment than those in a fully simulated environment.

show abstract

Autonomous Learning of Ball Trapping in the Four-Legged Robot League

Cited by 6 publications

References 5 publications

Development of an Augmented Environment and Autonomous Learning for Quadruped Robots

Development of an Augmented Environment and Autonomous Learning for Quadruped Robots

Learning Powerful Kicks on the Aibo ERS-7: The Quest for a Striker

Development of an Interactive Augmented Environment and Its Application to Autonomous Learning for Quadruped Robots

Contact Info

Product

Resources

About