I am writing my master thesis about Unmanned Aerial Vehicles and the sensor-camera data. It will be used for navigation and obstacle avoidance. Does anybody know any sources that I can download some sample uav data with Radar, Lidar, proximity, accelerometer, gyroscope, etc... sensors that would be useful for prediction of the next action such as turn right, move up. My aim is to analyise the input sensor and video data to predict the next move.
Add a comment
|
1 Answer
There are a few academic datasets that could be useful. The most common datasets are video though.
The Stanford Drone Dataset
https://cvgl.stanford.edu/projects/uav_data/
The Zurich Urban Micro Aerial Vehicle Dataset
http://rpg.ifi.uzh.ch/zurichmavdataset.html
Small UAV Position and Attitude, Raw Sensor, and Aerial Imagery Data Collected over Farm Field with Surveyed Markers This one contains raw sensor data in addition to aerial images. Probably the best option for you.
https://conservancy.umn.edu/handle/11299/170139?show=full
Columbus Surrogate Unmanned Aerial Vehicle (CSUAV) Dataset Overview
-
Most of these are survelliance video datasets from the down camera. I need ones from front camera and sensors also. They will be helpful for the device to decide where to move. Any other suggestions?– JeanCommented Apr 2, 2020 at 17:15
-
I'm not sure if you'll be able to find a dataset with sensors and camera with consistent angle of depression of 0 degrees. Most datasets have some angle of depression. It seems you will have to create your own dataset.– wacaxCommented Apr 10, 2020 at 13:23
-
Exactly. I can create my datasets from robot simulators, but the problem now is their sampling rates are too high so too much data is generated. However, it would be the best if I can find real datasets including a detailed sensing of the environment.– JeanCommented Apr 10, 2020 at 13:35