I recently attended The 2018 Society of Photographic Instrumentation Engineers (SPIE) Defense + Commercial Sensing event—the leading global sensing, imaging, and photonics technologies event. The conference was a mix of timely educational sessions, industry networking and leading-edge technology presentations, offering something for everyone. In this post, I want to take a closer look at three areas of interest to me: vehicle autonomy, deep learning and blockchain.
In the vehicle autonomy track, “Autonomous Systems: Sensors, Vehicles, Security and the Internet of Everything,” speakers discussed security, navigation and sensing and shared methods for preventing cyberattacks on autonomous cars and trucks. They explored safety enforcement, verification and validation, person detection and interpretation, sensing, cognitive architectures, cooperation, threat management and survivability and other related aspects. It’s clear that many issues have yet to be solved before autonomous vehicles become commonplace.
I presented my own thoughts on autonomy during the conference with my paper, “Finding Common Ground by Unifying Autonomy Indices to Understand Needed Capabilities.” It covered the definition of autonomy, compared autonomy with automation and described methods for measuring autonomy.
Obviously, autonomous sensing solutions are a hot topic, and companies are already selling partial solutions, such as automated sensors, artificially intelligent software and low SWAP hardware. I was impressed with the work they’re doing and see significant opportunities for KeyW to contribute. Autonomy is far from being a solved problem.
Panel discussions and papers on deep learning made it a popular conference topic. Discussions examined deep learning’s impact on the sensing community, and many papers described various applications of deep learning methods. Some of these papers covered electro-optical applications, where significant data sets are available. Sensing applications in other modalities, such as infrared, SAR and audio, are becoming possible as greater amounts of data are available in these modalities. Deep learning is providing solutions that weren’t possible with previous methods.
As you can imagine, blockchain was also an important topic. Various papers covered trust, ledger-partitioning methods and protecting high-performance computing systems. I’m hoping the larger community becomes better aware of this technology and applies it to secure sensor systems.
Still, the sensing community hasn’t completely woken up to blockchains. KeyW has an opportunity to adapt blockchain technologies to sensor products before the competition becomes aware of the potential.
Email us if you’d like to continue the conversation on the SPIE conference or any of the topics raised here.