We aim to create a smart, safe, swarm MAV. There are at least 7 sensors used in our MAV platform with a high-performance computation board like Odroid XU4, NUC. Mainly, the platform is used to verify our navigation, location and path planning algorithms in several kinds of environment.
MAV Platform
The third version MAV paktform takes Odroid XU4 as the computation board.
Block diagram:
The fourth version paltform enhance the computaion board with NUC board. And many vision algorithms can be run realtime in this board. This paltform is a plug and play platform which means this platform can be used both on MAV and groud robot.
The foruth version’s block diagram:
Sensor List
Product model of all sensors used in our datasets are listed:
| Sensor | Product Model | Collection frequency |
| Optical flow | Px4flow v1.3.1 | 20 |
| Stereo camera | 640×480×2 OV7725 | 30 |
| IMU | MPU9250 | 40 |
| Ultrasonic | MaxSonar MB1242 | 15 |
| RGB-D Camera | ASUS Xtion Pro Live | 40 |
| Lidar | Hokuyo UST-20LX | 40 |
| Lidar | Hokuyo UST-20LX | 40 |
| Vicon | Vero 360 | 100 |
| RTK GNSS receiver | Ublox M8P | 10 |
Environments
Our datasets are collected in four main scene:5m×5m×2.5m testing room with Vicon ’ Room’, 8m×12m×5m hall of office with Vicon in part of it ’ Hall’, 20m×20m open square outdoors ’ OutSquare’, and 50m2 area near the building outdoors ’ OutBuilding’.
Flight trajectories of datasets in indoor and outdoor scene are shown as following figure. where left one for indoor and right one for outdoor.
