Meno: | Tomáš
|
---|
Priezvisko: | Janeta
|
---|
Názov: | Planning smooth and safe arm movements using reinforcement learning
|
---|
Vedúci: | prof. Ing. Igor Farkaš, Dr.
|
---|
Rok: | 2023
|
---|
Kľúčové slová: | robotic arm, reinforcement learning, neural network, path plan- ning
|
---|
Abstrakt: | Reinforcement learning is a machine learning area, studying optimal strategy for an
agent in an environment with the possibility of taking various actions. Robotic path
planning is a well-known problem in which we are trying to find an effective path for a
robot that avoids all the obstacles. Classical algorithms for robotic path planning rely
on graph algorithms and are not effective enough in complex and dynamic environments
with many degrees of freedom. The thesis aims to implement a reinforcement learning
environment for a simulated robotic arm and train and test it using standard RL
algorithms.
|
---|