Deep reinforcement learning for high precision assembly tasks.
Tadanobu InoueGiovanni De MagistrisAsim MunawarTsuyoshi YokoyaRyuki TachibanaPublished in: IROS (2017)
Keyphrases
- high precision
- reinforcement learning
- high recall
- function approximation
- high reliability
- reinforcement learning algorithms
- model free
- learning algorithm
- deep learning
- state space
- achieve high precision
- neural network
- control problems
- policy search
- continuous state
- temporal difference learning
- robotic control
- data sets
- markov decision processes
- optimal control
- optimal policy
- direct policy search
- real time
- real world
- genetic algorithm
- multi agent reinforcement learning
- decision making
- stochastic approximation
- knowledge base
- case study
- learning agent
- hidden markov models
- dynamic programming
- high accuracy
- temporal difference