MIT researchers have unveiled a new way for consumers and businesses to take to the skies — by designing and developing their own drones.
The program, created by a team from the Massachusetts Institute of Technology (MIT)’s Computer Science and Artificial Intelligence Laboratory (CSAIL), is a prototype computer system (.PDF) which allows users to create a variety of drones.
According to the lab, users can design, simulate and build custom drones, altering the size, shape, and structure depending on how they want their drones fabricated and their purpose — which impacts on factors including payload tolerance, flight time, and battery life.
To demonstrate the system, CSAIL created a range of unusual drones, including a five-rotor “pentacopter” and a rabbit-shaped “bunnycopter.”
MIT says the program could pave the way for businesses and consumers to develop new drones able to perform a wider variety of functions.
As noted by the chief of the project, MIT professor Wojciech Matusik, one day there may no longer need to be a “one size fits all approach,” and rather, people looking to capitalize on drones for their business can tailor their devices without requiring an in-depth knowledge of electronics, physics, and design elements.
The interface itself allows users to design drones with different rotors and rods and guarantees that the fabricated end result will be able to take off, hover and land.
If you are able to design less “conventional” drones, however, they could be used for more unusual purposes — such as carrying objects with irregular shapes.
“For example, adding more rotors generally lets you carry more weight, but you also need to think about how to balance the drone to make sure it doesn’t tip,” says PhD student Tao Du.
“Irregularly-shaped drones are very difficult to stabilize, which means that they require establishing very complex control parameters,” Du added.
The new system removes the need for specialized knowledge by pulling data from databases full of drone parts, required specifications including payload, cost and battery usage, and using computer metrics to design the drone based on these parameters.
Elements including rod lengths and motor angles are automatically calculated to determine whether the design will work.
In addition, MIT researchers have added what they call an “alternating direction method” which reduces the number of drone design variables — fixing some and optimizing the rest — to reduce the number of factors which could negatively impact the drone’s performance.
“Once you decouple these variables, you turn a very complicated optimization problem into two easy sub-problems that we already have techniques for solving,” says Du.
The project was supported by the National Science Foundation, the Air Force Research Laboratory and the European Union’s Horizon 2020 research and innovation program. The system is currently only in the prototype stage, but it is hoped that eventually, the program will become open-source.
Earlier this month, MIT researchers revealed the creation of a new computer model which can give artificial intelligence systems a more ‘human’ and accurate way of recognizing faces.