|
Unmet Need
Real-time control and edge decision systems often depend on linear or quadratic optimization. Digital solvers can be too slow and energy-intensive for embedded uses such as autonomous vehicle control and medical classification. A practical hardware platform must also remain programmable so it can solve different optimization problems.
USC...
|
|
Background
Accurate detection of cracks in concrete and pavement infrastructure is critical for maintaining safety and durability, yet traditional manual inspections are labor-intensive and subjective. Deep learning models offer promise but are hindered by the scarcity of large, diverse, and well-annotated datasets. Existing datasets are limited in...
|
|
Market Opportunity
Next-generation wearable health-monitoring electronics and neural interfaces demand devices that are compliant, low-power, and intelligent. Conventional, rigid electronics lack the flexibility and stretchability needed to conform to the dynamic shapes and movements of the human body, limiting comfort and performance.
USC Solution
USC...
|
|
Background
Modern compression systems use a set of fixed mathematical tools to remove unnecessary information from each video frame. While these tools work well, they treat most video content in the same way and cannot adapt to the unique characteristics of each scene. As video resolutions, frame rates, and streaming demands continue to increase,...
|
|
Background
Current battery management systems, such as those in electric vehicles, grid storage and consumer electronics, can only estimate bulk metrics like capacity and resistance, which miss key degradation mechanisms. While electrochemical impedance spectroscopy (EIS) provides detailed insight into battery health, it requires specialized hardware...
|
|
Variance-weighted neural similarity analytics delivering noise-corrected, reproducible population spike comparisons.
Abstract
As neural recording systems have expanded in channel density and data complexity, reliably comparing population-level spike activity across trials, sessions, and subjects has remained a significant challenge. Traditional...
|
|
Background
To enable fault-tolerant quantum computing and reliable quantum simulations for drug development and materials science, quantum hardware must move beyond "noisy" prototypes to commercial-grade stability. Current benchmarking standards often miss systematic coherent errors that accumulate quadratically, creating a hidden bottleneck...
|
|
Market Opportunity
All electronic devices are susceptible to computational errors due to radiation effects. As electronics get smaller, they are more vulnerable to computational errors due to radiation effects. In mission-critical environments such as military, finance and healthcare, these computational errors are not tolerable. The best solution...
|
|
Market Opportunity
Demand for autonomous vehicles is expected to reach 29 million units by 2035. High-resolution detection and ranging technologies, such as LiDAR, are used in autonomous vehicles for remote sensing. Using LiDAR to create, steer, and detect high-resolution optical beams in free-space allows capturing real time 3D data for a diverse...
|
|
Quantum computing
Cryogenic quantum chip to classical world quantum interconnect for heterogeneous quantum devices (memory, sensors, detectors, circulators, qubit sources)
Abstract
USC inventors have designed a heterogeneously integrated quantum chip optoelectronics interposer (QuIP). QuIP enhances qubit performance, offers controlled coupling between...
|