Drone Modular Smart Pallet (DroneMSP)

To support the advancement of NASArsquo;s Unmanned Aircraft Systems (UAS) technologies, specifically in the areas of: (a) verification, validation, and certification and (b) sensing, perception, cognition, decision making, American GNC Corporation (AGNC) and California State University, Northridge (CSUN) are proposing a new technology referred to as a Drone Modular Smart Pallet (DroneMSP). This smart pallet consists of a reconfigurable sensor suite, flexible interfacing unit, processing with SD-card memory, and power management. The components are housed in a small form-factor and lightweight frame that can be easily attached to and detached from different vehicles. This smart pallet is designed to be plug-and-play for use on low-cost, common commercial drones, instantly granting them with the smart capabilities of multi-modality sensing with data acquisition and online sensor fusion processing. This technology will instantly enable NASA scientists and many other researchers to test and deploy their own algorithms and sensors on commercial drones. The collected data can be input into in-flight processing algorithms but will also be saved in public repositories to facilitate research by diverse groups with the ultimate goal of advancing Urban Air Mobility (UAM) and the testing of technologies as needed for unmanned flight in the National Airspace. For demonstrating the utility of the smart pallet, an object recognition and collision avoidance Use Case is included which shows how the sensor suite can provide data to an algorithm to conduct a task of relevance to UAM. Key innovations include: (1) plug-and-play hardware and software; (2) flight optimized design; (3) embedded cognition with obstacle avoidance and (4) data labeling scheme for sensor quality generation.