Robust computing and communications

Hybrid Cognitive-Neural Perception HRL is leading a team to develop a fully integrated, long-range, persistent, accurate threat-detection and cueing system that will provide advanced situational analysis to warfighters. This system integrates bio-inspired attention algorithms with brain-in-the-loop neural electroencephalography (EEG) signature detection of threats running on low-power neuromorphic hardware to yield exceptional detection performance over a 120°-by-20° field of view.

Neuro-Inspired Control Systems Using a unique 26 degrees-of-freedom robot shoulder, arm and hand, HRL researchers are developing neuro-inspired control systems that demonstrate fault tolerance and robustness under variable conditions. Like humans, who use perception-learning-action, or PLA cycles to accomplish vision-to-motor control for tasks such as reaching, we have created an innovative PLA application that can precisely control the robot arm under previously unforeseen conditions with minimal training. This level of adaptability and fault-tolerance exceeds the capability of traditional control algorithms.

Ultra-Wideband (UWB) Synchronization In a wireless communication system, it is important for a receiver to have accurate timing information for incoming data so that it can synchronize with the sender. The synchronization process is particularly difficult in impulse-IWB (I-UWB) systems because of the short pulse duration and the existence of multiple data paths. HRL has developed a coherent signal detection method that speeds up the synchronization process while keeping circuit complexity low. In addition, our approach allows the same circuit to be used for two functionalities at different phases of the receiving process—synchronization and demodulation.