Glossary
A
Action: A ROS 2 communication pattern for long-running tasks that provides feedback during execution and can be canceled.
Agent: An autonomous entity that perceives its environment and takes actions to achieve goals.
B
Behavior Tree: A hierarchical structure used to organize and execute robot behaviors in a logical manner.
C
Cognitive Architecture: A framework for organizing AI systems that models human cognition and decision-making.
D
DDS (Data Distribution Service): Middleware protocol and API standard for real-time, scalable, distributed data exchange.
Deep Learning: A subset of machine learning using neural networks with multiple layers to model complex patterns.
E
Embodied Intelligence: AI systems that interact with and operate within the physical world through robotic bodies.
F
Framework: A reusable set of libraries or APIs that provide generic functionality for building applications.
G
Gazebo: A physics-based simulation environment for robotics research and development.
H
Hardware Abstraction Layer: Software layer that allows higher-level code to interact with hardware without device-specific code.
I
Interface: A shared boundary across which two systems communicate and exchange information.
L
Learning Rate: A hyperparameter that controls how much to change the model in response to the estimated error each time the model weights are updated.
LiDAR: Light Detection and Ranging - a remote sensing method that uses light in the form of a pulsed laser to measure distances.
M
Middleware: Software that provides common services and capabilities to applications beyond what's offered by the operating system.
Model Predictive Control (MPC): An advanced method of process control that uses a model of the system to predict future behavior.
N
Navigation Stack: A collection of ROS packages that provide robot navigation capabilities.
Neural Network: A series of algorithms that endeavors to recognize underlying relationships in a set of data through a process that mimics how the human brain operates.
NVIDIA Isaac: A GPU-accelerated robotics development platform for building and deploying AI-powered robots.
P
Perception: The ability of a robot to interpret sensory information from its environment.
Physical AI: Artificial intelligence systems that interact with and operate within the physical world.
Publisher: A ROS 2 node that sends messages to a topic.
R
Real-time: Systems that process data and respond to inputs within a guaranteed time frame.
Robot Operating System (ROS): A flexible framework for writing robot software that provides services designed for a heterogeneous computer cluster.
ROS 2: The second generation of the Robot Operating System with improved real-time support, security, and multi-robot capabilities.
S
Service: A ROS 2 communication pattern that provides synchronous request-response interaction.
Simulation: The imitation of the operation of a real-world process or system over time.
Subscriber: A ROS 2 node that receives messages from a topic.
T
Topic: A ROS 2 communication channel where nodes publish and subscribe to messages.
Transformer: A deep learning model architecture based on the attention mechanism for processing sequential data.
V
Vision Language Action (VLA) Models: Multimodal AI models that integrate visual perception, language understanding, and physical action capabilities.
Visual Servoing: A technique that uses visual feedback to control the motion of a robot.
W
Waypoint: A location that marks a change in direction, a decision point, or a stopping point.