|
Using of Reinforcement Learning for Four Legged Robot Control
Ondroušek, Vít ; Březina, Tomáš (advisor)
The Ph.D. thesis is focused on using the reinforcement learning for four legged robot control. The main aim is to create an adaptive control system of the walking robot, which will be able to plan the walking gait through Q-learning algorithm. This aim is achieved using the design of the complex three layered architecture, which is based on the DEDS paradigm. The small set of elementary reactive behaviors forms the basis of proposed solution. The set of composite control laws is designed using simultaneous activations of these behaviors. Both types of controllers are able to operate on the plain terrain as well as on the rugged one. The model of all possible behaviors, that can be achieved using activations of mentioned controllers, is designed using an appropriate discretization of the continuous state space. This model is used by the Q-learning algorithm for finding the optimal strategies of robot control. The capabilities of the control unit are shown on solving three complex tasks: rotation of the robot, walking of the robot in the straight line and the walking on the inclined plane. These tasks are solved using the spatial dynamic simulations of the four legged robot with three degrees of freedom on each leg. Resulting walking gaits are evaluated using the quantitative standardized indicators. The video files, which show acting of elementary and composite controllers as well as the resulting walking gaits of the robot, are integral part of this thesis.
|
|
Robot path planning by means of reinforcement learning
Veselovský, Michal ; Liška, Radovan (referee) ; Dvořák, Jiří (advisor)
This thesis is dealing with path planning for autonomous robot in enviromenment with static obstacles. Thesis includes analysis of different approaches for path planning, description of methods utilizing reinforcement learning and experiments with them. Main outputs of thesis are working algorithms for path planning based on Q-learning, verifying their functionality and mutual comparison.
|
| |
| |
| |
| |
| |
| |
|
Stochastic policy in Q-lerning used for control of AMB
Březina, Tomáš ; Krejsa, Jiří ; Věchet, S.
A great intention is lately focused on Reinforcement Learning (RL) methods. The article is focused on improving model free RL method known as Q-learning used on active magnetic bearing model. Stochastic strategy and adaptive integration step increased the speed of learning approximately hundred times. Impossibility of using proposed improvement online is the only drawback, however it might be used for pretraining and further fined online.
|
|
Learning based control system of four-legged robot
Březina, Tomáš ; Houška, P. ; Singule, V.
Possible discretization technique of the continuous state space of four-legged robot using simultaneous compositions of behaviors is described in the contribution. Compositions are generated by the instances of two basic controllers. The aim is to automatically develop the gait policy. Possible composition strategies are implemented through undeterministic state machine. In the machine design stage the number of both states and transitions could be essetially reduced.
|