Competition: Dec. 14th, 2019
NeurIPS 2019
Vancouver, B.C. Canada<\/p>\n
Useful Links<\/h3>\n
Race Environments & Guidelines (opens in new tab)<\/span><\/a> Game of Drones is a\u00a0NeurIPS 2019 competition (opens in new tab)<\/span><\/a>\u00a0with the goal to push the boundary of building competitive autonomous systems through head-to-head drone races.<\/p>\n The teams will race a quadrotor drone in simulation through a series of gates against an opponent. There will be two drones on the same track at the same time that are allowed to block each other. The goal is to finish the race quicker than the opponent.<\/p>\n The competition uses AirSim (opens in new tab)<\/span><\/a>, which is an open-sourced simulation platform for robots. AirSim (opens in new tab)<\/span><\/a> was created to help accelerate research in deep reinforcement \/ imitation learning, computer vision and robotics for autonomous vehicles.<\/p>\n Previous drone race challenges, such as the IROS autonomous drone race challenge or the AlphaPilot challenge only need you to race against the clock.\u00a0 In contrast, here, you\u2019ll have to race\u00a0against an opponent on the same track.\u00a0<\/em>This means you can block your opponent if you are ahead and you have to overtake him when you are behind creating a whole range of interactions that your policy has to handle.<\/p>\n Besides being fun, this competition provides a fair comparison for the current state of the art.<\/p>\n If your current focus is either more on the planning or perception side, you\u2019ll most probably learn a thing or two. The audience is expected to be large \u2014 NeurIPS is the largest AI conference after all, so you can expect to get some attention if your algorithm scores high.<\/p>\n Prizes will also be awarded to the highest performing teams \u2013 we will announce them in the following weeks!<\/p>\n Register your team here (opens in new tab)<\/span><\/a>. Please make sure that your team\u2019s registration form is complete before submitting. Once submitted via your registration form, accepting any changes to registration details will be at the sole discretion of the organizers.<\/p>\n AirSim environments and detailed guidelines for the competition are available here (opens in new tab)<\/span><\/a>.<\/p>\n *New*<\/span> The submission instructions (opens in new tab)<\/span><\/a> are now available.<\/p>\n *New*<\/span> The leaderboard (opens in new tab)<\/span><\/a> is now active. (Validation round).<\/p>\n *New*<\/span> The Qualification Round will start Oct 15.<\/p>\n\n\n\n\n\n The organizing team behind the game of drones are the developers of AirSim from Microsoft Research and robotics researchers from Stanford who share the vision of a simulator realistic enough to provide a training and testing environment for autonomous mobile robots such as drones or self-driving cars.<\/p>\n Many people are helping to make this happen, but the core team consists of:<\/p>\n For any inquiries please see our FAQ section or send us an email<\/a>.<\/p>\n\n\n\n\n\n \n Register your team here (opens in new tab)<\/span><\/a>. Please make sure that your team\u2019s registration form is complete before submitting. Once submitted via your registration form, accepting any changes to registration details will be at the sole discretion of the organizers.<\/p>\n<\/p><\/div>\n \n We will be publishing the detailed procedure by July 31st 2019.<\/p>\n<\/p><\/div>\n \n We will be releasing reference baselines for all the three tiers by July 15th, 2019. You can use these reference baselines as an optional starting point.<\/p>\n<\/p><\/div>\n \n Airsim clients can be run on Windows and Linux. The reference baselines will be in Python.<\/p>\n<\/p><\/div>\n \n Yes, we encourage academic publications relating to the challenge. If you do decide to publish results, please cite AirSim platform as:<\/p>\n \n Prizes will be awarded to the highest performing teams – we will announce them in the following weeks!<\/p>\n<\/p><\/div>\n \n For any inquiries please see our FAQ section or send us an email<\/a><\/p>\n<\/p><\/div>\n<\/p><\/div>\n<\/p>\n\n\n\n\n\n Drone Kit<\/p>\n<\/div>\n Fat Shark Dominator HDO FPV Goggles<\/p>\n<\/div>\n<\/div>\n<\/p><\/div>\n<\/p><\/div>\n<\/p>\n\n\n","protected":false},"featured_media":610338,"template":"","meta":{"msr-url-field":"","msr-podcast-episode":"","msrModifiedDate":"","msrModifiedDateEnabled":false,"ep_exclude_from_search":false,"_classifai_error":"","msr_hide_image_in_river":0,"footnotes":""},"msr-opportunity-type":[187426],"msr-region":[256048],"msr-locale":[268875],"msr-program-audience":[243724],"msr-post-option":[],"msr-impact-theme":[],"class_list":["post-593767","msr-academic-program","type-msr-academic-program","status-publish","has-post-thumbnail","hentry","msr-opportunity-type-challenges","msr-region-global","msr-locale-en_us","msr-program-audience-students"],"msr_description":"","msr_social_media":[],"related-researchers":[],"tab-content":[{"id":0,"name":"About the competition","content":"Registration for the contest has now closed.<\/strong>\r\n\r\nCompetition: Dec. 14th, 2019\r\nNeurIPS 2019\r\nVancouver, B.C. Canada\r\n
<\/b>*New*<\/span><\/strong> S (opens in new tab)<\/span><\/a>ubmission Instructions (opens in new tab)<\/span><\/a>
*New* <\/span><\/strong>Leaderboard (opens in new tab)<\/span><\/a>
<\/strong>AirSim (opens in new tab)<\/span><\/a><\/p>\nPush the boundaries of competitive autonomous systems<\/h2>\n
![\"Game](\"https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2019\/06\/final-airsim-01.gif\")
<\/h3>\nWhat\u2019s special about this competition?<\/h3>\n
\n
\nThe competition comes in three tiers of different difficulty and level of autonomy.<\/h3>\n
\n
Given: Ground-truth poses of your and your opponent\u2019s drone.
Challenge: Finish the track quicker than your opponent, without crashing.<\/li>\n
Given: Camera images (RGB from the front camera) and ground-truth pose of your drone.
Challenge: Finish the track as quick as possible, without crashing.<\/li>\n
Given: Camera images (RGB from the front camera) and ground-truth poses of your but not your opponent\u2019s drone.
Challenge: Finish the track quicker than your opponent, without crashing.<\/li>\n<\/ul>\n<\/li>\n<\/ul>\nWhy should I participate\/care about the game of drones?<\/h3>\n
\n
\nWe’ll release our submission as an opponent in the online interface to benchmark against. Note that the code will not be made available.<\/li>\n
\nAll submissions made in the timeframe above will be considered for qualification.<\/p>\n\n
\nLast possible change to the reference submission, to allow participants enough
\ntime to adjust their algorithms.<\/li>\n
\nThe detailed qualification rules will be released, specifying what exactly has to be passed to be qualified for the placement rounds.<\/li>\n
\nAnnouncement of qualified teams for the live challenge, invitation for poster presen-
\ntation.<\/li>\n
\nAll present qualified teams participate in the live competition.<\/li>\n<\/ul>\n\n\n\n\n\nWho is behind this?<\/h3>\n
\n
\n \tHow do I register? <\/h4>\n
\n \tHow do I submit my entry to the challenge? <\/h4>\n
\n \tDo you have some example code? <\/h4>\n
\n \tWhat platforms \/ languages can I use? <\/h4>\n
\n \tCan I publish results of my experiments? <\/h4>\n
@inproceedings{airsim2017fsr,<\/pre>\nauthor = {Shital Shah and Debadeepta Dey and Chris Lovett and Ashish Kapoor},<\/pre>\ntitle = {AirSim: High-Fidelity Visual and Physical Simulation for Autonomous Vehicles},<\/pre>\nyear = {2017},booktitle = {Field and Service Robotics},<\/pre>\neprint = {arXiv:1705.05065},<\/pre>\nurl = {https:\/\/arxiv.org\/abs\/1705.05065} }<\/pre>\n<\/p><\/div>\n\n \tWhat can I win? <\/h4>\n
\n \tWhere can I get help? <\/h4>\n
Prizes for Game of Drones<\/h2>\n
Category 1<\/h3>\n
\n
Category 2<\/h3>\n
\n
Category 3<\/h3>\n
\n
![\"Drone](\"https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2019\/06\/drone-kit-300x272.jpg\")
(opens in new tab)<\/span><\/a><\/p>\n![\"Fat](\"https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2019\/06\/fat-shark-hdo-fpv-goggles_2-300x215.jpg\")
(opens in new tab)<\/span><\/a><\/p>\nUseful Links<\/h3>\r\nRace Environments & Guidelines<\/a>\r\n<\/b>*New*<\/span><\/strong> S<\/a>ubmission Instructions<\/a>\r\n*New* <\/span><\/strong>Leaderboard<\/a>\r\n<\/strong>AirSim<\/a>\r\n
Push the boundaries of competitive autonomous systems<\/h2>\r\nGame of Drones is a\u00a0NeurIPS 2019 competition<\/a>\u00a0with the goal to push the boundary of building competitive autonomous systems through head-to-head drone races.\r\n
<\/h3>\r\nThe teams will race a quadrotor drone in simulation through a series of gates against an opponent. There will be two drones on the same track at the same time that are allowed to block each other. The goal is to finish the race quicker than the opponent.\r\n\r\nThe competition uses AirSim<\/a>, which is an open-sourced simulation platform for robots. AirSim<\/a> was created to help accelerate research in deep reinforcement \/ imitation learning, computer vision and robotics for autonomous vehicles.\r\nWhat's special about this competition?<\/h3>\r\nPrevious drone race challenges, such as the IROS autonomous drone race challenge or the AlphaPilot challenge only need you to race against the clock.\u00a0 In contrast, here, you'll have to race\u00a0against an opponent on the same track.\u00a0<\/em>This means you can block your opponent if you are ahead and you have to overtake him when you are behind creating a whole range of interactions that your policy has to handle.\r\n
\r\n \t
The competition comes in three tiers of different difficulty and level of autonomy.<\/h3>\r\n
\r\n \t
Why should I participate\/care about the game of drones?<\/h3>\r\nBesides being fun, this competition provides a fair comparison for the current state of the art.\r\n\r\nIf your current focus is either more on the planning or perception side, you\u2019ll most probably learn a thing or two. The audience is expected to be large \u2014 NeurIPS is the largest AI conference after all, so you can expect to get some attention if your algorithm scores high.\r\n\r\nPrizes will also be awarded to the highest performing teams - we will announce them in the following weeks!\r\n\r\nRegister your team here<\/a>. Please make sure that your team\u2019s registration form is complete before submitting. Once submitted via your registration form, accepting any changes to registration details will be at the sole discretion of the organizers.\r\n\r\nAirSim environments and detailed guidelines for the competition are available here<\/a>.\r\n\r\n*New*<\/span> The submission instructions<\/a> are now available.\r\n\r\n*New*<\/span> The leaderboard<\/a> is now active. (Validation round).\r\n\r\n*New*<\/span> The Qualification Round will start Oct 15."},{"id":1,"name":"Schedule","content":"
\r\n \t
\r\n \t
Who is behind this?<\/h3>\r\nThe organizing team behind the game of drones are the developers of AirSim from Microsoft Research and robotics researchers from Stanford who share the vision of a simulator realistic enough to provide a training and testing environment for autonomous mobile robots such as drones or self-driving cars.\r\n\r\nMany people are helping to make this happen, but the core team consists of:\r\n
\r\n \t
@inproceedings{airsim2017fsr,<\/pre>\r\nauthor = {Shital Shah and Debadeepta Dey and Chris Lovett and Ashish Kapoor},<\/pre>\r\ntitle = {AirSim: High-Fidelity Visual and Physical Simulation for Autonomous Vehicles},<\/pre>\r\nyear = {2017},\r\n\r\nbooktitle = {Field and Service Robotics},<\/pre>\r\neprint = {arXiv:1705.05065},<\/pre>\r\nurl = {https:\/\/arxiv.org\/abs\/1705.05065} }<\/pre>\r\n[\/panel]\r\n\r\n[panel header=\"What can I win?\"]\r\nPrizes will be awarded to the highest performing teams - we will announce them in the following weeks!\r\n[\/panel]\r\n\r\n[panel header=\"Where can I get help?\"]\r\n\r\nFor any inquiries please see our FAQ section or send us an email<\/a>\r\n\r\n \r\n\r\n[\/panel]\r\n\r\n[\/accordion]"},{"id":4,"name":"Prizes","content":"[row]\r\n\r\n[column class=\"m-col-12-24\"]\r\nPrizes for Game of Drones<\/h2>\r\n
Category 1<\/h3>\r\n
\r\n \t
Category 2<\/h3>\r\n
\r\n \t