Research efforts include the exploration of privacy-preserving light-fields for occupancy and activity sensing methods; integration of robotic, autonomous and self-regulating system controls; and the development of algorithms and simulation tools. LESA ‘controls’ methodologies are focused on fundamental pattern recognition such as Bayesian and neural networks, dynamic clustering, and system information disseminate across communication networks.
LESA’s concept of light transport analysis and plenoptic light-field sensing using digitized spectral reflectance and Time of Flight (TOF) or ‘Indoor LIDAR’ for activity mapping, is a transformative and powerful approach to address some of the toughest enginnering challenges in the connected environment. The TOF approach is superior to other occupancy tracking systems (including RF reflection/attenuation sensing) in terms of overall value, cost, benefit and precision. The integrated sensor platforms are developed adaptive and providing rich datasets for machine learning and artificial intelligence enabling embedded autonomous adaptive control of lighting and building systems, such as HVAC – and funded in part by a DOE ARPA-E award.