ML Collision Avoidance via 5G MEC Edge Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current collision avoidance systems face challenges in effectively tracking multiple vehicles and handling various scenarios in real-time, leading to potential delays in alerting drivers of impending collisions, which can result in accidents.
Innovation Solution
A collision avoidance system utilizing a trained machine learning model that receives data from on-board diagnostics, cameras, and traffic light controllers, integrated with a Multi-access Edge Computing (MEC) network and 5G NR wireless signals to provide low-latency alerts to vehicles, enabling rapid collision detection and prevention.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional collision avoidance systems process telematics information, then collision detection capability is provided, but processing delays occur leading to late driver alerts
Solution Approach 1:
The patent transitions collision detection from vehicle-centric onboard processing to network-centric cloud-based processing using 5G NR and MEC networks. This dimensional shift in processing architecture enables parallel processing of multiple vehicle telemetry streams, reducing individual vehicle processing delays while maintaining comprehensive collision detection coverage across the network.
Solution Approach 2:
The patent introduces 5G NR wireless communication and MEC network infrastructure as intermediary layers between vehicles and collision detection algorithms. These intermediaries enable high-speed data transmission and distributed computing, reducing the time required to process telematics information and generate collision alerts compared to traditional direct onboard processing.
2Adaptability or versatility
If multiple vehicles are tracked simultaneously, then comprehensive collision detection is achieved, but system complexity increases
Solution Approach 1:
The patent creates a universal collision detection system where a single cloud-based platform performs multiple functions: tracking multiple vehicles, processing diverse telematics data types, detecting various collision scenarios, and providing alerts. This multi-functional approach simplifies individual vehicle systems while achieving comprehensive multi-vehicle monitoring through the shared network infrastructure.
Data Source
AI summary
A device may include a memory storing instructions and processor configured to execute the instructions to receive information relating to a plurality of vehicles in an area. The device may be further configured to use a trained machine learning model to determine a likelihood of collision by one or more of the plurality of vehicles; identify one or more relevant vehicles of the plurality of vehicles that are in danger of collision based on the determined likelihood of collision; and send an alert indicating the danger of collision to at least one of the identified one or more relevant vehicles.