|
|
Using of Reinforcement Learning for Four Legged Robot Control
Ondroušek, Vít ; Maga,, Dušan (referee) ; Maňas, Pavel (referee) ; Singule, Vladislav (referee) ; 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.
|
|
|
Implementaion of the controllers of a mobile walking robot
Krajíček, Lukáš ; Věchet, Stanislav (referee) ; Ondroušek, Vít (advisor)
This diploma thesis deals with design and implementation of the controllers of a mobile walking robot. The advantage of these controllers are their kinematics and geometrics independent representation, which allow to use them for different robot types and tasks. In this thesis the contact controller is designed, which minimizes residual forces and torques at the robot's center of gravity, and thereby stabilize robot's body. Next the thesis deals with a posture controller, which maximizes a heuristic posture measure to optimize posture of robot body. Because of this optimization, legs are moved away from their limits and therefore they have more working space for next move. Implementation of the chosen solution is made on the robot's MATLAB mathematical model. Controllers are composed into a control basis, that allows to solve general control tasks by simultaneous combination of contained controllers. The algorithm was created for that simultaneous activation and its operation was explained on flow charts.
|
|
|
The Study of Controllers for Four-legged Walking Robots
Klvaňa, Roman ; Věchet, Stanislav (referee) ; Ondroušek, Vít (advisor)
This work is based on literary search and is focused on description of various approaches in walking gait generation for four-legged walking robots. There are six various groups of aprroaches described in this thesis. Main functional principles as well as one representative robot of each approach is described. The summarize of advantages and disadvanteges of each approach is used to apoint an appropriate control approach for robot Quasimodo walking gait generation in the conclusion chapter of this bachleor thesis.
|
|
|
Design of Model of a Walking Autonomous Mechanism and its Simulation of Movements in Difficult Terrain
Szabari, Mikuláš ; Jánoš,, Rudolf (referee) ; Bobovský,, Zdenko (referee) ; Knoflíček, Radek (advisor)
The first part of the work is devoted to the current state, which discusses the design and step cycles of legged mechanisms and also the resistant terrain. The following second part critically evaluates the current state and defines the solved problem. The third part sets specific goals and creates a work procedure. The solution of the work follows by fourth part, which realizes the morphological analysis of the structure of the legged mechanism and the analysis of the resistant terrain in which the mechanism moves. The fifth part is devoted to the creation of models of legged mechanisms, its initial test and the choice of a specific construction of a legged mechanism for resistant terrain. The following sixth part simulates the movement of the legged mechanism and deals with its optimization. The simulation is confirmed by the experiment performed in the seventh part. The end of the work is devoted to contributions, discussion of results and conclusion.
|
|
|
Using of Reinforcement Learning for Four Legged Robot Control
Ondroušek, Vít ; Maga,, Dušan (referee) ; Maňas, Pavel (referee) ; Singule, Vladislav (referee) ; 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.
|
|
|
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.
|
|
|
Implementaion of the controllers of a mobile walking robot
Krajíček, Lukáš ; Věchet, Stanislav (referee) ; Ondroušek, Vít (advisor)
This diploma thesis deals with design and implementation of the controllers of a mobile walking robot. The advantage of these controllers are their kinematics and geometrics independent representation, which allow to use them for different robot types and tasks. In this thesis the contact controller is designed, which minimizes residual forces and torques at the robot's center of gravity, and thereby stabilize robot's body. Next the thesis deals with a posture controller, which maximizes a heuristic posture measure to optimize posture of robot body. Because of this optimization, legs are moved away from their limits and therefore they have more working space for next move. Implementation of the chosen solution is made on the robot's MATLAB mathematical model. Controllers are composed into a control basis, that allows to solve general control tasks by simultaneous combination of contained controllers. The algorithm was created for that simultaneous activation and its operation was explained on flow charts.
|
|
|
The Study of Controllers for Four-legged Walking Robots
Klvaňa, Roman ; Věchet, Stanislav (referee) ; Ondroušek, Vít (advisor)
This work is based on literary search and is focused on description of various approaches in walking gait generation for four-legged walking robots. There are six various groups of aprroaches described in this thesis. Main functional principles as well as one representative robot of each approach is described. The summarize of advantages and disadvanteges of each approach is used to apoint an appropriate control approach for robot Quasimodo walking gait generation in the conclusion chapter of this bachleor thesis.
|
|
|
Using Q-learning for four-legged robot walking control
Březina, Tomáš ; Houška, P. ; Singule, V.
Significant goal in walking robot design is realization of autonomous system which is capable of motion in unknown environment. One of the possibilities how to reach necessary adaptability of control system without modeling of complex and unpredictable cases of system behaviour is the application of machine learning. From various learning paradigms the Q-learning is particularly attractive.
|
|
| |
guest ::
