SPECIAL NOTICE
99 -- Multilateral Autonomy Prize Challenge $50,000
- Notice Date
- 8/9/2023 1:28:52 PM
- Notice Type
- Special Notice
- Contracting Office
- NIWC ATLANTIC NORTH CHARLESTON SC 29419-9022 USA
- ZIP Code
- 29419-9022
- Solicitation Number
- N65236_SNOTE_01E4B2E6
- Archive Date
- 11/16/2023
- Point of Contact
- Giancarlo Dumenigo, Contract Specialist, Phone: 843-218-2375
- E-Mail Address
-
giancarlo.dumenigo.civ@us.navy.mil
(giancarlo.dumenigo.civ@us.navy.mil)
- Description
- The NIWC Atlantic Palmetto Tech Bridge (PTB) is seeking innovative technical approaches for 3 different problem statements. Participants can submit up to one technical approach per topic. 1. Computer Vision Surprise Challenge Artificial Intelligence (AI) methods for detection and classification of objects of interest in imagery or video is extremely important for Naval Intelligence, Surveillance, and Reconnaissance (ISR), self-driving cars, X-ray analysis, digital inspection in manufacturing, and other applications. Most methods to perform this task are achieved using supervised machine learning, where a set of labeled training data is employed to optimize a model�s parameters to improve accuracy. However, what happens when these models in operation see novel objects or conditions that were not seen during training? This competition is aimed at finding innovative techniques for handling these situations for image detection and/or classification. We are specifically looking for solutions that can best account for one or more of four problems: (1) the background environment has significantly changed from the training set (for example, weather or lighting conditions), (2) a different camera is being used during operation than during training, (3) new types of objects or classes now need to be identified that weren�t present during training, and (4) the character of the object itself has changed such that it may be harder to identify (e.g., a sticker was placed on a STOP sign that slightly changes its appearance, such that a human can still identify it as a STOP sign easily but an algorithm may encounter difficulty). Additionally, if these detected objects could also be authenticated using advanced image classification technologies that would also be of interest. 2. Risky Facility Location Challenge Given a limited number of facilities that can be built, we wish to know the best location to place each facility to meet geographical demands, costs, capacities, and transportation distances. An example of this problem in practice would be a company trying to determine where to place a new manufacturing plant or distribution facility to maximize their return on investment. We wish to identify approaches to solve this problem with a twist: what happens if some of the potential geographic areas are known for having high crime, and therefore, there is a risk that sometimes the warehouse or supply vehicles may get robbed or broken into? Or if some areas often experience roadwork or construction that makes shorter distances occasionally longer? Other risk factors could include predicted climate related impacts and/or workforce market characteristics. We are looking for innovative ideas for tackling this Capacitated Facility Location (CFL) problem with these �risk� or �uncertainty� twists. 3. Multi-Sensor Fusion for Autonomous Driving Self-driving cars, aircraft, or watercraft may rely on different types of sensors such as light detection and ranging (LiDAR), electro-optical (EO), and infrared (IR) to be able to operate effectively and respond to their environments (e.g. collision avoidance, safety, navigation, etc.). Each of these or other sensors provides different information to a self-driving system about their environment. A particular challenge then is to develop methods that relate this information together or can fuse it to present a common operating picture of the environment. Individuals or teams will present their ideas during an Innovation Pitch Jam in September to a panel to be evaluated. Find more details by finding the Prize Challenge at challenge.gov. Submission period: Phase 1 open until 10/31/23 10:00 PM EDT
- Web Link
-
SAM.gov Permalink
(https://sam.gov/opp/be2e0177e464480a84aaee0e8c6762dd/view)
- Record
- SN06782702-F 20230811/230809230046 (samdaily.us)
- Source
-
SAM.gov Link to This Notice
(may not be valid after Archive Date)
| FSG Index | This Issue's Index | Today's SAM Daily Index Page |