Reinforcement learning (RL) is a general and well-known method that a robot can use to learn an optimal control policy to solve a particular task. We would like to build a versatile robot that can learn multiple tasks, but using RL for each of them would be prohibitively expensive in terms of both time and wear-and-tear on the robot. To remedy this problem, we use the Policy Gradient Efficient ...

In this paper path planning and obstacle avoidance for a statically stable biped robot using PRM and reinforcement learning is discussed. The main objective of the paper is to compare these two methods of path planning for applications involving a biped robot. The statically stable biped robot under consideration is a 4-degree of freedom walking robot that can follow any given trajectory on fla...

Goalkeeper robot in the league of humanoid robots has a key role so that its wrong decision and incorrect jump can cause changes in the result of a competition. This paper presents a method to detect the ball on the ground during the match, determine the coordinates of the ball with respect to the robot, calculating the speed and direction of the ball and eventually the time and place that the ...

I n this paper. we present the mechanism, system cor$guration, basic control algorithm and integrated ,functions of the Nonda humanoid robot. Like its human counterpart, this robot lzas the ability to move forward and backward, sideways to the right or the left, as well as diagonally. I n addition, the robot can turn in any direction, walk up and down stairs continuously. Furthermore, due to it...

In this thesis, we consider the problem of an autonomous mobile robot operating in a stochastic reward field to maximize total rewards collected in an online setting. This is a generalization of the problem where an unmanned aerial vehicle (UAV) collects data from randomly deployed unattended ground sensors (UGS). Specifically, the rewards are assumed to be generated by a Poisson point process....

This work presents a combination of a teach-and-replay visual navigation and Monte Carlo localization methods. It improves a reliable teach-and-replay navigation method by replacing its dependency on precise dead-reckoning by introducing Monte Carlo localization to determine robot position along the learned path. In consequence, the navigation method becomes robust to dead-reckoning errors, can...

The present paper investigates an intuitive way of robot path planning, called robot teaching by demonstration. In this method, an operator holds the robot end-effector and moves it through a number of positions and orientations in order to teach it a desired task. The presented control architecture applies impedance control in such a way that the end-effector follows the operator’s hand with d...

Bipedal walking is a difficult task due to its intrinsic instability and developing successful controller architectures for this mode of locomotion has proved substantially more difficult than for other types of walking (Beer et al. 1998). This type of walking has been tackled from different directions that all of them have been divided to two major approaches of static walking (Kato et al. 197...

Mobile robot navigation is one of the basic problems in robotics. In this paper, a new approach is proposed for autonomous mobile robot navigation in an unknown environment. The proposed approach is based on learning virtual parallel paths that propel the mobile robot toward the track using a multi-layer, feed-forward neural network. For training, a human operator navigates the mobile robot in ...

Design, modeling, manufacturing and experimental analysis of a six degree freedom robot, suitable for industrial applications, has been described in this paper. The robot was designed on the assumption that, each joint has an independent DC motor actuator, with gear reduction and measuring sensor for angular joint position. Mechanical design of the robot was done using Mechanical Desktop and ma...