Learning by observation with mobile robots: a computational approach

Dixon, Kevin R.; Khosla, Pradeep K.

doi:10.1109/robot.2004.1307136

Cited by 22 publications

(13 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Each linear dynamical system is used to derive a convergent controller, with a small region around the final state considered its goal. The algorithm can be run online and was used in conjunction with several other methods to build a mobile robot system that performed LfD by tracking a human user [Dixon and Khosla, 2004b]. This system differs from CST in three ways.…”

Section: Related Workmentioning

confidence: 99%

Robot learning from demonstration by constructing skill trees

Konidaris

Kuindersma

Grupen

et al. 2011

The International Journal of Robotics Research

262

191

View full text Add to dashboard Cite

We describe CST, an online algorithm for constructing skill trees from demonstration trajectories. CST segments a demonstration trajectory into a chain of component skills, where each skill has a goal and is assigned a suitable abstraction from an abstraction library. These properties permit skills to be improved efficiently using a policy learning algorithm. Chains from multiple demonstration trajectories are merged into a skill tree. We show that CST can be used to acquire skills from human demonstration in a dynamic continuous domain, and from both expert demonstration and learned control sequences on the uBot-5 mobile manipulator.

show abstract

Section: Related Workmentioning

confidence: 99%

Robot learning from demonstration by constructing skill trees

Konidaris

Kuindersma

Grupen

et al. 2011

The International Journal of Robotics Research

262

191

View full text Add to dashboard Cite

show abstract

“…The robot is then able to make a decision to execute one of 4 motion primitives(unit actions) based on its sensory readings. In [11] the robot uses a laser sensor to detect and recognize objects of interest. A policy is learned to predict subgoals associated with the detected objects rather than directly predicting the motion primitives.…”

Section: Navigationmentioning

confidence: 99%

Deep Active Learning for Autonomous Navigation

Hussein

Gaber

Elyan

2016

Communications in Computer and Information Science

View full text Add to dashboard Cite

“…The LfD literature may be divided into two categories: those which learn plans [22,31] and those which learn (usually stateless) policies [3,19] (for stateful examples see [8,13]). In most cases, the plan literature builds sparse machines describing occasional changes in behavior, whereas many, but not all, policy methods learn fine-grained changes in action, such as might be found in trajectory planning or control.…”

Section: Related Workmentioning

confidence: 99%

Real-Time Training of Team Soccer Behaviors

Sullivan

Luke

2013

RoboCup 2012: Robot Soccer World Cup XVI

View full text Add to dashboard Cite

Abstract. Training robot or agent behaviors by example is an attractive alternative to directly coding them. However training complex behaviors can be challenging, particularly when it involves interactive behaviors involving multiple agents. We present a novel hierarchical learning from demonstration system which can be used to train both single-agent and scalable cooperative multiagent behaviors. The methodology applies manual task decomposition to break the complex training problem into simpler parts, then solves the problem by iteratively training each part. We discuss our application of this method to multiagent problems in the humanoid RoboCup competition, and apply the technique to the keepaway soccer problem in the RoboCup Soccer Simulator.

show abstract

Learning by observation with mobile robots: a computational approach

Cited by 22 publications

References 6 publications

Robot learning from demonstration by constructing skill trees

Robot learning from demonstration by constructing skill trees

Deep Active Learning for Autonomous Navigation

Real-Time Training of Team Soccer Behaviors

Contact Info

Product

Resources

About