Turning Intersection Radio Messages into ADAS Decisions

A V2I radio simply delivers bytes. The engineering challenge sits one layer up: inside the vehicle's compute stack, software must parse SAE J2735 ASN.1 payloads, reconcile signal timing against noisy GNSS position, and hand a trustworthy intersection state to the ADAS planner in milliseconds.

The Application Layer That Sits Between the V2I Radio and the ADAS Planner

The automotivev2i.com domain covers what happens after a Roadside Unit's SPaT or MAP frame lands in the vehicle's network buffer: ASN.1 UPER decoding, timestamp and coordinate frame reconciliation, confidence scoring against onboard perception, and delivery of a clean intersection-state object to ADAS modules such as Green Light Optimal Speed Advisory and Intersection Movement Assist.

Core Engineering Areas

• SPaT/MAP Message Parsing and Decoding — Roadside Units broadcast SPaT and MAP messages using the compact Unaligned Packed Encoding Rules (UPER) defined by SAE J2735. Before any ADAS module can reason about a red light or a permitted turn lane, in-vehicle software must run a validated ASN.1 decoder, verify message integrity, and normalize the result into a stable internal schema.
• ADAS Fusion of Signal-Phase Countdown Timers — Once a SPaT frame is decoded, the countdown values are probabilistic, not deterministic — actuated signal controllers can extend a phase based on live traffic. The ADAS fusion layer must therefore treat minEndTime and maxEndTime as a confidence interval, continuously re-estimating an optimal approach speed as fresh SPaT updates arrive at up to 10 Hz.
• MAP Message Fusion with Onboard Lane-Level Localization — MAP messages describe intersection geometry as a graph of Lanes and Connections referenced to a single reference point in absolute latitude/longitude. The onboard localization stack, by contrast, tracks the vehicle relative to a locally cached HD-map tile. Fusing the two requires a coordinate transform pipeline resilient to GNSS drift of several meters in dense urban canyons.

Technical Articles

• Building a Production SPaT Decoder for Green Light Optimal Speed Advisory
• Fusing MAP Messages with Onboard Localization for Intersection Movement Assist

Related Cluster Architecture Specifications

• AutomotiveSDV — Full SDV Stack & Reference Architecture
• A2HCard — Agent-to-Human (A2H) Interface Cards & In-Cabin Autonomous Interaction Manifests
• A2HumanCard — Standardized In-Vehicle AI Agent Capability Discovery, Trust & Authorization Cards
• AIDrivenSynBio — AI-Driven Synthetic Biology for Automotive In-Cabin Air Purification, Bio-Engineered Pathogen Scavengers & Molecular Biosensors