Visual Inertial Simultaneous
Hey guys and welcome back so in this lecture we can be talking about VISLAM also known as Visual-Inertial. Simultaneous Localization and Mapping it is a technology that is powering augmented reality experiences on a wide variety of devices and forms part of the framework in the thevuforia Fusion API SLAM to understand, how VISLAM works we first need to understand what SLAM.
So imagine that you are in a new city trying to find your way around a completely unknown place, what would you do first you to look around to find familiar buildings markers or signs and once you recognize a familiar landmark you can then figure out where you are in relation to those landmarks.
Recognize any Landmarks
Now, what happens if you do not recognize any landmarks at all well you’ll be declared as lost obviously, however if you take the time to observe the environment. The more landmarks you’ll recognize, and begin to pull a mental image or map of the place that you’re into do you agree that you may have to navigate the certain environment several times before you start becoming familiar with a previously unknown location. How does it relate to a robot
Now, in the same way, a robot with SLAM capabilities tried to map an unknown environment while simultaneously trying to figure out where it is at, also known as localization. The challenge comes from doing both at the same time The robot needs to know its position before answering the question of what the environment looks like. It also has to figure out where it is at without the benefit of already having a map.
Hence, simultaneous localization and mapping or SLAM is a way of solving this problem using specialized equipment and algorithms. For this process to work the robot or device must have exceptional Odometry performance. Odometry, Odometry is the use of data from motion sensors to estimate changes in position over time.
It is used in robotics by some late and wheeled robots to estimate their position relative to a starting location. Now at any sort of measurement instrument, there’s only a small margin of error or noise with odometry readings. These errors are taken into account in various algorithms that also remap the area to make up for this deficiency.
So in a nutshell slam tackles two basic tasks: 1) mapping what does the world looks like. 2) localization Where am I VISLAM now that we understand what SLAM is. Now, what is “VI” in the SLAM glad you asked they are Visual and Inertial sensors that are used for slam functionality some can work with a multitude of sensors but on your smartphone,
The only way it can map its surroundings for augmented reality applications is by using the camera and its IMU for Inertial Navigation for the visual element is a lot of methods that can be used for both estimation but they all essentially try to do feature matching some common algorithms that you may or may not be familiar with our SURF, SIFT, Difference of Gaussian, ORB amongst many other techniques. Vuforia VISLAM.
Vuforia uses VISLAM
How does it relate to augmented reality? Speaking of augmented reality Vuforia uses VISLAM in his before fusion engine which enables features such as a device tracker which provides six degrees of freedom device to pose ground plane allowing virtual content to be placed on horizontal planes in the environment and extend the tracking enabling extended tracking for all Vuforiafor target types VISLAM.
Is an algorithm implemented by Vuforiacombining the benefits of visual-inertial odometry (VIO) and slam. Vuforia VISLAM has several benefits it works better in low feature environments compared to slam-based tracking it provides an estimate of the scale of the world ie the device tracker motion estimate will correspond to the real-world motion and it provides robustness to recover when tracking is completely lost over.
Vo only solutions when using Ground Plane before s first land solution will attempt to estimate the distance between the camera and the ground plane to provide a smooth and user experience before an engine will use the initial hints to estimate skill based on the height to the flow provided via the developer API the default of this hint is 1.4 meters the average height of an adult holding a device.In the hands in some cases, you can see the scale estimation at work in the initialization phase,
App and this is when augmentation is corrected in the apparent size when placed onto the ground plane initialization so VISLAM is just one of the algorithms for air applications you’re probably familiar with arkitfrom Apple and arcore from Google If available the fusion API will automatically use ARKIT or arcore if this platform enables are no available on the user’s devices to automatically use the slam if the device has required sensors and has been calibrated by Vuforia otherwise if the device cannot support that it will use SLAM.
On all the devices okay so that is it from me Okay, so that is it from me I hope you enjoyed this video, and now that you understand what SLAM is and our VISLAM is being used for augmented reality applications and you’ll find it as part of the portfolio fusion API if you are interested in learning more about augmented reality.