AIDA is said to be purpose-built for a flexible range of ITS applications, making each user-selected function completely automated and cost-efficient.
ARCADIA, Calif. — LILIN introduces the AIDA traffic management platform, with embedded artificial intelligence (AI) that learns, adapts and detects the details of street-level vehicle activity.
Engineered for seamless integration into LILIN Windows-based VMS products including Navigator Control Center, Navigator Corporate and Navigator Enterprise, AIDA installs on the edge and uses metadata to bring real-time traffic data to organizations and municipalities.
AIDA is said to be purpose-built for a flexible range of ITS applications, making each user-selected function automated and cost-efficient. For example, the platform can extract data from vehicles traveling on multilane highways or urban streets, allowing municipalities to better enforce parking and traffic violations, recover stolen cars or assist law enforcement in Amber Alerts.
AIDA can recognize 12 license plates per second on vehicles traveling up to 125 miles per hour, according to the company. Police can store and later access this data, enabling investigators to locate a suspect’s whereabouts and behavioral patterns. Besides monitoring cars, trucks and motorcycles, the platform can detect pedestrians loitering or entering restricted areas.
In a parking lot, AIDA can compare license plate data to pre-defined lists of allowed or excluded vehicles before taking an appropriate action, such as opening a gate or generating an alert if an unauthorized vehicle enters an area as a value-added service. It can also locate available parking slots and broadcast this information to incoming cars.
In addition to its security role, AIDA can capture vital operational data about parking infrastructures, including the number of cars served by time, day and length of stay. This data can then be applied to help optimize a staffing schedule and maximize the profitability of a parking lot’s rate structure.
AIDA is said to remove the hassles of traditional computer vision techniques that require extensive, CPU-draining image processing. In contrast, the platform’s software incorporates an AI engine that is instructed to learn what to look for and is able to run multiple models on distributed AI CPU, GPU and VPUs.