Artificial intelligence invites itself in drone races

Having become unbeatable in the game of go, chess, and even Starcraft, the AI will soon embark on racing drones to fully compete autonomously with a human champion in a competition called  Artificial intelligence robotic racing. The closing date of the registration of autonomous drone development teams is getting closer, the tension is rising ...

Confronting completely autonomous drones to the champions of driving racing drones is the goal of the Artificial intelligence robotic racing (AIRR) organized by the  Drone Racing League (DRL). As Futura had already said, the project supported by aircraft manufacturer Lockheed Martin and Nvidia is called AlphaPilot, which is reminiscent of the name given to Google's AI, AlphaZero, AlphaGo ... It is open to all drone enthusiasts or specialists in artificial intelligence (AI).

In the coming months, the teams will have to develop an AI associated with automatic learning algorithms so that the drone can evolve completely autonomously. To date, more than 300 teams have already registered. Registration and qualifications take place from the HeroX platform. The deadline to participate is March 8th. The selections will be made throughout the month of April. In any case, the common hardware base for designing the drone is Nvidia's Jetson AGX platform, which is dedicated to autonomous machines. The subtlety is that it is not possible to pre-program navigation. The drones are thus delivered to themselves during their race.

From virtual to real

At first, the teams will fly their autonomous drones in a virtual way on circuits also boosted to AI and inspired video games. It is also a team from MIT that developed the simulator that will be used for qualifying. This racing season will begin August 1st. At the end of this one, the fastest freestyle drone will face the 2019 Allianz World Champion DRL in real-world conditions.

The challenge for AlphaPilot teams will be to successfully transpose their virtual inventions into the real world. At the end of the adventure, there is a nice carrot with over 2 million dollars in cash for the most advanced and fastest AI. It remains to be seen if, on a real circuit, the AI will perform better than the best racing drone driver of the moment.

WHAT YOU MUST REMEMBER

• The drone races are a new development ground for artificial intelligence.
• By launching the AlphaPilot program in conjunction with the Drone Racing League and Nvidia, Lockheed Martin hopes to emerging technologies that will serve him in his defense and aerospace activities. 

AlphaPilot: UAV and AI pilots will compete in a real race

Article by Marc Zaffagni published on 11/09/2018

Launched by Lockheed Martin, the  Drone Racing League, and NVidia, AlphaPilot is a drone racing project that will pit human pilots against artificial intelligence. For the defense and aerospace giant, the challenge is to prepare the technologies of tomorrow.

Starting next year, humans will face artificial intelligence (AI) during drone races. Called AlphaPilot, this project is the result of a rather surprising combination between Lockheed Martin, the  Drone Racing League (professional drone league) and NVidia. The competition is endowed with two million dollars in prizes, including a bonus of $ 250,000 to the first team whose autonomous drone will manage to beat a human driver.

The races will take place as part of a new format launched by the  Drone Racing League named  Artificial intelligence robotic racing (AIRR). AlphaPilot is open to all drone enthusiasts or specialists in artificial intelligence, whether students, researchers or employees of a company. They will have to create an AI and a machine learning algorithm capable of flying a drone on the different routes proposed without prior programming of the navigation.

Participants will be provided with all the necessary equipment so that everyone has the same basic equipment. In particular, they will need to use NVIDIA 's Jetson computing platform to create embedded artificial intelligence applications for the automotive, robotics and drones. In its statement, Lockheed Martin said that its engineers will play the role of "  mentors " for the teams involved.

A way for Lockheed Martin to do R & D

"  By participating in this competition, your knowledge and ideas can contribute directly to the future of autonomous transportation, delivery, disaster relief, and even space exploration, " says the company specializing in defense and aerospace. In summary, AlphaPilot is a way for Lockheed Martin to do R & D, find future talent and advance the technologies that will equip its products of tomorrow.

Creating an autonomous drone capable of competing with the reflexes and dexterity of human pilots is obviously not an easy task. But the experiment has already been tried with some success. Last year, the  Jet Propulsion Laboratory of NASA organized a race between autonomous drones and a human pilot. If the latter had emerged with several seconds apart, drones driven by an AI had demonstrated a great consistency in their trajectories. However, as for the game of go, chess, and video games, human unpredictability and adaptability will be the major obstacles that the machine will have to overcome to overcome.