Parrot has introduced an open development kit for the design of advanced applications for autonomous navigation, obstacle avoidance, indoor navigation and 3D mapping for drones and other robotic platforms.
According to the company, the Parrot S.L.A.M.dunk integrates advanced software applications based on a simultaneous localization and mapping algorithm, also known as S.L.A.M.
Parrot says S.L.A.M. technology enables the drone to understand and map its surroundings in 3D and to localize itself in environments with multiple barriers and where GPS signals are not available.
With an embedded NVIDIA Tegra K1 processor, Parrot S.L.A.M.dunk allows the implementation of algorithms of computer vision and autonomous navigation that are optimized for NVIDIA GPU. In addition, the Parrot S.L.A.M.dunk is designed to minimally impact the battery life of a drone.
Additionally, the development kit integrates a fish-eye stereo camera with a 1500 x 1500 resolution at 60 fps, inertial measurement unit, ultrasound sensor, magnetometer and barometer.
Parrot says it developed S.L.A.M.dunk, which includes an HDMI port, to be as easy and user-friendly as possible. Sensors and algorithms are fully integrated in order to speed the design of advanced applications for drones, and all Ubuntu functionalities and benefits from the robot operating system framework are embedded in the software.
Weighing less than 4.9 oz (140 g), Parrot S.L.A.M.dunk fits on various types of Linux-based drones and robotic platforms, including quadcopters and fixed-wing drones or rolling robots.
Once placed, Parrot S.L.A.M.dunk should be linked to the battery of its host drone by a 3.5 mm jack cable and a USB cable so that it can send and receive information and commands to and from the drone.
Developed by robotic researchers for robotic researchers, says Parrot, the kit is a ready-to-use module that will accelerate the development of tomorrow’s applications for drones and robotic platforms.
The software will be available in the fourth quarter of this year.