Often a robot is first tested in a simulation before being

used in a real scenario

A robot cannot improve learning skills in manipulation tasks

A robot makes decisions in an environment to maximize a

cumulative reward

A robot is always directly used in the real scenario without

the need for simulated training

A simple version of an actuator

with 6 degrees of freedom that can interact with the physical environment

An end effector that can interact

with the physical environment

a programmable structure

responsible for interacting with and effectuating change to the physical

environment.

A simple autonomous robot that

can interact with the physical environment

Gazebo

Webots

Choregraphe 

Rviz

A robot is an embodied agent that can

potentially develop behavioural and cognitive skills by relying on the input

provided by their sensors and a fitness (or reward) function.

A Robot is an embodied agent that can

integrate information over time and store in the body or in the relation

between their body and the environment. 

A robot is an AI agent where internal

layers of a  ANNs /  Deep learning architecture represents the

only way to extract features.

A robot is a particular AI agent that

can only use features extracted through actions with the external

environment. 

B because there is no instrument without error however small,

and at all times there will be an estimation error from localisation, an

estimation error of the landmarks and an error in the robot’s controls.

B because even if assuming a perfect ego-estimate of the

robot’s location, there can still be an

A because using a GPS we can precisely estimate the position

of the robot and we know where the robot is located at all times.

Either A or B, depending on the circumstances in the

environment.