MSU Technologies

Robust Adaptive Modular Perception System

VAlue proposition

Neural network-based algorithms have advanced the state-of-the-art in many perception-related tasks such as detection, tracking, and prediction, which are critically important for autonomous vehicles, ADAS, and other related technologies. However, integrating disparate neural network-based algorithms together can be challenging due to varying system-level requirements, non-standard network input and output formats, and custom network features. A framework for developing a node-based system for integrating datasets, perception algorithms, and other processing modules in a single cohesive graph-based system is desired.

Description of Technology

This technology is a modular perception system that avoids computationally expensive data transfers by relaying data between modules as shared data accelerator memory segments. The use of shared memory segments is unique compared to other existing systems, which generally require copying between accelerator and processor memory to maintain modularity. This system represents a novel framework architecture for constructing and evaluating perception pipelines.

Benefits