New Drone System Aims for Full Autonomy

A team of engineers from the University of Zurich has developed an open-source quadrotor drone for use as a test bed to accelerate research into agile, autonomous flight.

“Agilicious,” the platform, seeks to give the field’s researchers a leg up in their work by providing a standardized, modular platform to solve sophisticated flight problems commensurate with the technology’s evolution.

“Agile flight comes with ever-increasing engineering challenges because performing faster maneuvers with an autonomous system requires more capable algorithms, specialized hardware and proficiency in system integration,” states the research paper published in Science Robotics.

“This work aims to bridge this gap through an open-source agile-flight platform, enabling everyone to work on agile autonomy with minimal engineering overhead.”

Agilicious provides the compute power necessary for full autonomy on an airframe that’s small, lightweight and powerful enough for next-generation flight requirements.

The platform provides high thrust-to-weight and torque-to-inertia ratios for flight agility, as well as onboard vision sensors, graphics processing unit (GPU)-accelerated compute hardware, a real-time flight controller and a versatile software stack, the researchers said.

The drone itself is made from off-the-shelf parts, including brushless motors powering five-inch propellers.

The aircraft can handle agile flight at more than 5g acceleration and fly up to 80 miles per hour. Its design takes advantage of recent advances in motor, battery and frame construction from the first-person view racing community, according to the researchers.

The team conducted simulated and real-world tests of Agilicious to evaluate its capabilities as a test vehicle. In one trajectory-tracking experiment, the drone reached speeds of 35 miles an hour and accelerations of 4G. The drone was able to complete the course after several iterations of algorithm development and substantial tuning efforts.

“The transition from simulation to real-world deployment requires no source-code changes or adaptions, which reduces the risk of crashing expensive hardware and is one of the major features of Agilicious accelerating rapid prototyping.”

Autonomous, agile flight promises real-world applications, such as in time-critical missions including search and rescue, aerial delivery, flying cars and exploration.