"The Airpeak S1 is equipped with many sensors to provide maneuverability and safe flight. The most important of these is a 5-way stereo camera that captures information about the aircraft and its surroundings from images. In the "Airpeak S1", the vision sensing processor uses these 5 x 2 = 10 image sensors, and the IMU (Inertial Measurement Unit), which detects the angular velocity and acceleration of the aircraft, for a total of 11 sensors.
After all, I wanted to fulfill the user's request to fly the drone safely. Safety is ensured by detecting obstacles in the surroundings, and by performing highly accurate self-position estimation even under sudden aircraft movements and violent vibrations of the highly mobile "Airpeak S1", stable flight even in places where GNSS is difficult to enter. I wanted to be able to provide.
When thinking so, for example, if there are only two stereo cameras, there is a risk that the necessary information cannot be obtained depending on the location and environment. Since vision sensing with a stereo camera requires the presence of a textured object within the range of measurement, it is possible to detect surrounding obstacles by increasing the number of observable directions, and it also demonstrates the performance of self-position estimation. It will be easier to meet the conditions. However, it is not easy to install 5 stereo cameras. In addition to the installation location, cost, and weight issues I mentioned earlier, we need a high-performance microprocessor that has a large number of input systems and can process a large amount of video in real time."
No comments:
Post a Comment
All comments are moderated to avoid spam and personal attacks.