Editing Motion capture (section)

=== Robotics ===
Indoor positioning is another application for optical motion capture systems. Robotics researchers often use motion capture systems when developing and evaluating control, estimation, and perception algorithms and hardware. In outdoor spaces, it's possible to achieve accuracy to the centimeter by using the Global Navigation Satellite System ([[Satellite navigation|GNSS]]) together with Real-Time Kinematics ([[Real-time kinematic positioning|RTK]]). However, this reduces significantly when there is no line-of-sight to the satellites — such as in indoor environments. The majority of vendors selling commercial optical motion capture systems provide accessible open source drivers that integrate with the popular Robotic Operating System ([[Robot Operating System|ROS]]) framework, allowing researchers and developers to effectively test their robots during development.

In the field of aerial robotics research, motion capture systems are widely used for positioning as well. Regulations on airspace usage limit how feasible outdoor experiments can be conducted with Unmanned Aerial Systems ([[Unmanned aerial vehicle#Terminology|UAS]]). Indoor tests can circumvent such restrictions. Many labs and institutions around the world have built indoor motion capture volumes for this purpose.

Purdue University houses the world's largest indoor motion capture system, inside the Purdue UAS Research and Test (PURT) facility. PURT is dedicated to UAS research, and provides tracking volume of 600,000 cubic feet using 60 motion capture cameras.<ref>{{Cite web |title=Purdue's Home for Drone Systems Engineering |url=https://engineering.purdue.edu/AAE/Aerogram/2021-fall/articles/ac-very-unique-facility |access-date=2023-09-18 |website=Aerogram Magazine - 2021-2022 |language=en}}</ref> The optical motion capture system is able to track targets in its volume with millimeter accuracy, effectively providing the true position of targets — the "ground truth" baseline in research and development. Results derived from other sensors and algorithms can then be compared to the ground truth data to evaluate their performance.