2026-06-18 — views

Physical AI V2X 2026 — Waymo Sensor Fusion vs Tesla FSD Camera-Only: The Vehicle-to-Everything Smart City Connectivity Benchmark

Neither Waymo nor Tesla currently depends on V2X, but Tesla's 6M-vehicle fleet creates the largest cooperative perception opportunity if implemented.

Article 200 in the Physical AI Benchmark Series — Milestone Edition

This is article 200 in the Physical AI Benchmark Series — a milestone in long-form autonomous vehicle and robotics analysis. Previous articles in this series have covered the full spectrum of physical AI: Waymo commercial operations, Tesla FSD architecture, humanoid robots, HD mapping, AV safety data, regulatory timelines, investor frameworks, unit economics, and the global competitive landscape. Article 200 adds a new dimension: how autonomous vehicles integrate with the infrastructure of smart cities through Vehicle-to-Everything (V2X) connectivity.

V2X represents the next integration layer for AV technology — the point where autonomous vehicles stop being isolated sensor platforms and start participating in a connected urban information ecosystem. This benchmark covers the V2X technology landscape, the current V2X posture of both Waymo and Tesla FSD, global smart city deployment progress, and the 2028 outlook for AV-urban infrastructure integration.

Section 1 — The V2X Technology Landscape: Four Links and Two Competing Standards

V2X (Vehicle-to-Everything) connectivity is defined by SAE J2945 standard as encompassing four communication links, each addressing a distinct dimension of how vehicles interact with their environment.

The Four V2X Communication Links

V2X Link	What It Connects	Key Data Exchange	AV Relevance
V2V (Vehicle-to-Vehicle)	AVs communicate directly with each other	Position, speed, heading, intent data; enables cooperative maneuvers (merging, intersection crossing, platooning) without requiring infrastructure	High: enables cooperative perception — sharing detected objects between AVs; range est. 300m (est.) line-of-sight
V2I (Vehicle-to-Infrastructure)	AVs receive data from traffic signals, roadside units (RSUs), traffic management centers	SPaT (Signal Phase and Timing) broadcasts — real-time knowledge of signal timing before visual detection is possible; construction zone alerts; emergency vehicle preemption	Very high: intersection efficiency; enables GLOSA (Green Light Optimal Speed Advisory)
V2P (Vehicle-to-Pedestrian)	AVs receive data from pedestrian smartphones (DSRC or C-V2X phones)	Pedestrian crossing intent before pedestrian is visually detected	High at night: particularly important for night conditions where visual detection is harder
V2N (Vehicle-to-Network/Cloud)	AVs communicate with cloud-based traffic management, HD map update services, fleet coordination	Real-time traffic data, OTA software updates, fleet coordination	Already active: both Waymo (ROC connectivity) and Tesla (OTA updates, traffic data)

Two Competing V2X Radio Standards

The industry is divided between two radio technologies for V2X communication, and the competition has significant policy, regulatory, and deployment implications:

DSRC (Dedicated Short-Range Communications, 802.11p/WAVE):

The original US V2X standard; deployed in some US cities
5.9 GHz band; low latency (approximately 20ms); works without cellular network
Range est. 300–1,000m (est.)
Critical setback: The FCC’s 2020 ruling reallocated most of the 5.9 GHz band away from DSRC — a significant policy blow that has caused automotive industry momentum to shift away from DSRC

C-V2X (Cellular V2X, 3GPP Release 14 and later):

Uses LTE/5G cellular infrastructure for vehicle connectivity
Supports both direct mode (PC5 interface — no cellular network needed) and network mode (via cellular network)
C-V2X direct mode has similar range to DSRC but better non-line-of-sight performance
Widely adopted in China; EU moving toward C-V2X under ITS Delegated Act
Most AV companies (including Waymo and Tesla) are moving toward C-V2X because it leverages existing 5G infrastructure investments

US Policy Landscape

2021 Infrastructure Investment and Jobs Act: $110M allocated for V2X pilot deployments across US cities
USDOT SPaT Challenge: 20+ US cities committed to deploying SPaT broadcasts from traffic signals
National Roadway Safety Strategy (2022): V2X identified as a safety tool
NHTSA V2X rulemaking: Planning NPRM (Notice of Proposed Rulemaking) for V2X in new vehicles, expected 2025–2027
5.9 GHz band split: After 2020 FCC ruling, 45 MHz retained for V2X; 45 MHz reallocated to Wi-Fi

Global V2X Leadership

Region	V2X Status	Standard
China	Global leader — Shenzhen, Beijing, Shanghai, Chongqing deployed C-V2X on hundreds of km of urban roads; national standard GB/T 35356 mandates C-V2X in new passenger vehicles from 2027; “Smart Road, Smart Car, Smart City” initiative integrates C-V2X with AV pilots	C-V2X
EU	ITS Delegated Act (2023) requires C-V2X capability in new vehicles from 2026+ in EU; ETSI ITS standards define EU C-V2X protocols; Netherlands, Germany, France leading deployment	C-V2X
US	Building infrastructure via $110M IIJA program; 20+ cities with SPaT pilots; NHTSA NPRM pending; industry coalescing around C-V2X	Transitioning from DSRC to C-V2X
Japan	ITS Connect (DSRC-based) deployed in Tokyo, Nagoya, Osaka; transitioning to C-V2X	DSRC → C-V2X
South Korea	C-V2X deployed in K-City AV test facility; nationwide C-V2X roadmap by 2030	C-V2X

Section 2 — Waymo’s V2X Posture: Sensor Fusion First, V2X as Enhancement

Waymo’s current architecture is built around sensor fusion: LIDAR + camera + radar + HD maps. This stack provides sufficient situational awareness for commercial operations in current geofenced zones — without any V2X dependency. V2X, for Waymo, is an additive enhancement layer, not a current operational requirement.

V2X Dimension	Waymo Current Approach	V2X Enhancement Opportunity	Timeline
Traffic signal reading	Sensor stack reads traffic signal states in real-time via camera (multiple streams cover all intersection approaches); LIDAR detects signal housing position; HD map contains static signal locations + phase layouts; Waymo does not currently rely on V2I SPaT broadcasts	V2I SPaT integration would give Waymo advance knowledge of signal timing before visual detection is possible; enables GLOSA (Green Light Optimal Speed Advisory) — energy-efficient speed profiles approaching signals	Near-term: USDOT SPaT pilots provide infrastructure; Waymo could integrate V2I receiver in Gen 6 vehicle generation; pilot testing likely by 2027–2028 (est.)
Construction zone alerts	HD maps require manual updates when construction zones change road geometry; construction zones are among the most challenging scenarios for AVs (unexpected lane closures, flaggers, temporary signage, altered geometry)	V2I-based construction zone alerts via USDOT Work Zone Data Exchange (WZDx standard — already operational): real-time construction zone boundaries + lane status broadcast to connected vehicles; Waymo integration would reduce HD map update latency for construction zones	WZDx integration is near-term: data standard is published and operational; Waymo could integrate WZDx feeds into its map update pipeline (software change, not hardware)
Emergency vehicle preemption	Detects emergency vehicles via audio (sirens) and visual (lights); pulls over when detected; can struggle with obstructed audio or multiple simultaneous sirens	V2I emergency vehicle preemption: traffic signal preemption systems broadcast emergency vehicle approach data; Waymo receives advance warning before audio/visual detection is possible; enables proactive pull-over before emergency vehicle is within sensor range	Safety-critical use case regulators may prioritize; depends on traffic signal system having V2X preemption broadcast
V2V cooperative perception	Each Waymo vehicle operates as an independent sensor platform; no real-time perception data shared between vehicles in the same geofence	V2V cooperative perception: Waymo vehicles sharing real-time detected objects (pedestrians, cyclists, occluded vehicles) could significantly improve perception in dense urban environments; a pedestrian occluded from Vehicle A’s sensors might be detected by Vehicle B and shared via V2V	Longer-term enhancement: requires V2V hardware (C-V2X radio) not yet in Waymo’s Gen 5/6 sensors; communication protocol development + privacy-preserving object sharing design required; potential deployment: 2028–2030 (est.)
School zone alerts	School zone signage read visually; speed limit reductions handled via HD map + visual sign detection; school hours + pedestrian activity not detected in advance	V2I school zone activation systems broadcast when school zones are active (school hours, crossing guard active); Waymo integration enables proactive speed reduction before visual zone signage is reached	Near-term safety improvement; deployment depends on city V2I infrastructure investment
Waymo overall V2X posture	Operates effectively without V2X — sensor fusion provides full situational awareness for current commercial operations in geofenced zones	V2X integration roadmap sequence (est.): (1) V2N cloud data feeds already active (WZDx, traffic data); (2) V2I SPaT integration for energy efficiency; (3) emergency vehicle preemption V2I; (4) V2V cooperative perception as longer-term enhancement	V2X integration limited by: hardware (Gen 5/6 communication hardware) and availability of V2X infrastructure in operating cities

Waymo V2X context: Waymo’s constrained geofenced operating zones are actually an advantage for V2I integration — a single city’s infrastructure investment can cover Waymo’s entire operating area. Phoenix, Arizona and Austin, Texas (both Waymo operating cities) have active state DOT V2X programs, making them natural early candidates for Waymo V2I integration pilots.

Section 3 — Tesla FSD’s V2X Posture: Camera-Only with 5G Cloud Dependency

Tesla FSD operates on a pure camera-based neural network architecture. Unlike LIDAR-based systems, Tesla’s approach reads all environmental signals — including traffic lights, signs, and lane markings — from camera feeds processed by neural networks. V2X, for Tesla, would provide a parallel data channel that enhances FSD performance without requiring hardware sensor additions beyond a V2X radio chip.

V2X Dimension	Tesla Current Approach	V2X Enhancement Opportunity	Timeline
Traffic signal reading	FSD reads traffic signals via camera using neural networks; FSD v12/v13 classifies traffic signal states (red/yellow/green, arrow phases, pedestrian signals) from camera feeds; Traffic Light and Stop Sign Control (TLSSC) feature; does not currently receive V2I SPaT data	V2I SPaT integration provides advance signal timing knowledge before camera detection; particularly useful in low-visibility conditions (sun glare, rain obscuring camera view) or when signal is obstructed by large vehicles; SPaT = software addition plus V2I radio chip	Tesla could add V2I capability as hardware option in future vehicle generations; given preference for camera-only architecture, V2I integration likely only happens if regulatory requirements mandate it or safety data shows clear benefit
Speed limit compliance	FSD reads posted speed limit signs from cameras; Navigation on Autopilot uses map data (TomTom + crowd-sourced detection) for speed limit compliance; FSD v13 improved sign recognition accuracy	V2I dynamic speed limit broadcasts (VSL — variable speed limit systems on highways) could provide real-time speed limit data, bypassing potential camera-based sign misreading in poor conditions; also relevant for work zone speed limit changes	Tesla already integrates map-sourced speed limit data alongside camera detection; V2I dynamic speed limits would be an additional data source
Construction zone navigation	FSD relies on camera-based detection of construction zone markings (orange cones, signage, temporary lane markings) plus OTA map updates; FSD v12/v13 neural net approach handles construction zone geometry changes better than earlier rule-based systems (generalizes from visual scene)	WZDx work zone data exchange integration: same as Waymo, Tesla could integrate real-time construction zone boundary data into its map update pipeline; camera-only FSD with WZDx would have advance knowledge of construction zone boundaries before visual detection	WZDx integration is a software addition (API integration, not hardware); Tesla’s OTA update capability means this could be deployed fleet-wide rapidly once developed
V2V cooperative perception	Tesla vehicles do not share real-time perception data between vehicles; each Tesla operates as an independent perception platform; Tesla’s data flywheel collects training data from the fleet but this is batch/async, not real-time V2V	Tesla’s 6M+ FSD-capable vehicle fleet operating in the same geographic areas creates a theoretical opportunity for V2V cooperative perception at a scale no other AV operator can match; if even 1% of Tesla vehicles shared real-time detected object data, the cooperative perception mesh would cover most US urban areas	Long-term opportunity (2028–2032 est.): requires real-time V2V communication protocol, privacy-preserving object data sharing, latency requirements for safety-critical cooperative perception
V2N cloud connectivity	Tesla vehicles have strong V2N connectivity: real-time traffic data integration, OTA updates (FSD software updates over cellular), Sentry Mode cellular uploads; built-in LTE/5G modems in recent generations	Integration of additional smart city data feeds (event management, road incident alerts, dynamic routing from traffic management centers) could improve FSD navigation efficiency in congested urban areas	Near-term: Tesla Navigation already integrates real-time traffic; adding city traffic management center API integration would be a software enhancement
Tesla Robotaxi smart city integration	Tesla’s Austin Robotaxi service will need to integrate with Austin’s traffic management systems (signal preemption, geofenced zone alerts, event management) as the fleet scales; Austin has active V2X pilot programs	Tesla’s Austin Robotaxi could be a test platform for V2X integration in a commercial AV context; Texas’s permissive AV regulatory environment and Austin’s smart city initiatives could accelerate V2X-Robotaxi integration	Near-term test case: Austin smart city V2X integration with Tesla Robotaxi could inform Tesla’s broader V2X strategy

Tesla V2X context: Tesla’s nationwide consumer fleet geography creates a fundamentally different V2X challenge than Waymo’s geofenced zones. Deploying V2I coverage across the full US road network where Tesla vehicles operate is not tractable — Tesla’s V2I benefit depends on concentration of vehicles in specific urban corridors where infrastructure investment is sufficient. Tesla’s V2N (cloud connectivity) strength is the more tractable near-term V2X advantage.

Section 4 — Global Smart City AV Integration: China Leading, EU Following, US Building

The global V2X deployment landscape has a clear hierarchy: China is the most aggressive deployer, the EU is building a regulatory mandate framework, and the US is investing in pilot infrastructure while regulatory mandates remain pending.

Geography	V2X Deployment Status	AV Integration	2026 Milestone
China (C-V2X leader)	Largest C-V2X deployment globally: Shenzhen (60+ km C-V2X equipped roads), Beijing (100+ km), Shanghai, Chongqing; national standard GB/T 35356 mandating C-V2X in new vehicles from 2027; government-funded “Smart Road, Smart Car, Smart City” initiative; C-V2X RSUs at thousands of intersections across major cities	BYD, Nio, Xpeng, Li Auto all deploying C-V2X capable vehicles; local AV operators (Apollo/Baidu, WeRide, AutoX) integrating V2X into AV operations; government-mandated cooperation between smart road operators and AV fleet operators	China surpasses 2M C-V2X capable vehicles on road (est.); first C-V2X-mandatory vehicle models from domestic OEMs; government AV operating zones in major cities require V2X capability
EU (ITS Delegated Act)	EU ITS Delegated Act (2023): C-V2X capability required in new vehicles from 2026+ in EU; ETSI ITS standards define EU C-V2X protocols; EU member states deploying C-V2X infrastructure on trans-European road network (TEN-T); Netherlands, Germany, France leading EU V2X deployment	Waymo’s planned EU expansion would require V2X compliance for EU market entry; Tesla’s EU manufacturing (Berlin Gigafactory) gives Tesla incentive to develop EU-compliant V2X from the production line	EU V2X-capable vehicle requirement enters force for new EU type-approved vehicles; first EU V2X corridors on TEN-T highway network operational
US (Building infrastructure)	$110M from Infrastructure Investment and Jobs Act for V2X pilots (2022–2026); USDOT SPaT Challenge operational in 20+ cities; WZDx standard fully operational; NHTSA V2X NPRM expected 2025–2027; 5.9 GHz band split: 45 MHz retained for V2X; automotive industry coalescing around C-V2X	Waymo’s operating cities (SF, Phoenix, LA, Austin) have varying V2X infrastructure levels; Austin (TX DOT V2X pilot) and Phoenix (AZ DOT V2X program) more active; SF/LA have USDOT SPaT pilots but limited coverage	NHTSA V2X NPRM publication expected; first US city with citywide C-V2X SPaT broadcast coverage; Waymo V2X integration in at least one commercial market
Japan / South Korea	Japan: ITS Connect (DSRC-based) deployed on major urban arterials in Tokyo, Nagoya, Osaka; transitioning to C-V2X for next generation. South Korea: C-V2X deployed in K-City AV test facility; nationwide C-V2X roadmap by 2030	Waymo has no current Japan/Korea operations; Tesla has sales in both markets; Korea’s C-V2X roadmap relevant for Tesla’s South Korean FSD expansion	Limited direct impact on Waymo/Tesla benchmark in 2026; relevant for medium-term international AV expansion planning

Key 2026 policy event: The EU ITS Delegated Act’s C-V2X requirement for new EU type-approved vehicles represents the first major regulatory mandate requiring V2X hardware as standard equipment. This will affect Tesla’s EU manufacturing (Berlin Gigafactory produces for the EU market) and creates a compliance deadline that will accelerate V2X hardware integration into Tesla’s vehicle production line by 2026–2027.

Section 5 — V2X and Smart City Benchmark Scorecard

Dimension	Waymo	Tesla FSD	Edge	2028 Outlook
Current V2X dependency	Low: operates without V2X; sensor fusion stack provides full situational awareness for current operations	Low: operates without V2X; camera-based signal reading + map-sourced traffic data sufficient for current operations	Roughly equal (neither currently depends on V2X; both operate effectively without it)	V2X becomes increasingly valuable as urban AV operations scale; neither company is V2X-dependent today
V2X enhancement benefit	High: HD map + V2I integration creates a two-layer verification system for traffic signal states + construction zones; SPaT enables energy-efficient GLOSA across Waymo’s geofenced zones	Moderate: V2I SPaT provides parallel signal timing data to camera-based reading; most valuable as a redundancy/backup channel; camera already handles signal reading adequately in good conditions	Waymo (higher marginal benefit from V2X because V2I adds a sensor modality Waymo currently lacks; LIDAR/camera detect signals but V2I provides timing data that no sensor can match)	Both benefit; Waymo’s constrained geofenced zones are easier to deploy V2I coverage than Tesla’s nationwide geography
V2V cooperative perception opportunity	Small fleet (est. 2,500 vehicles in 4 cities): V2V cooperative perception would cover limited geographic area	Large fleet (6M+ FSD-capable vehicles nationwide): V2V cooperative perception at scale would create the densest cooperative perception mesh of any AV program globally	Tesla (scale advantage in V2V cooperative perception if implemented)	Tesla’s V2V cooperative perception opportunity is one of the most significant untapped advantages in the Physical AI race; implementation is architecturally complex but the potential is enormous
China smart city integration	Not yet relevant (Waymo has no China operations)	Active: Tesla China fleet already in smart city context; regulatory pressure to comply with China C-V2X standards; China data center established	Tesla (already operating in China’s V2X-integrated smart city environment)	Tesla’s China V2X compliance experience will inform its global V2X strategy
Regulatory V2X compliance	Waymo’s EU expansion will require V2X compliance (ITS Delegated Act); US NHTSA V2X mandate could affect Waymo’s fleet hardware; Waymo’s Gen 6 vehicle design must accommodate V2X hardware	Tesla’s EU manufacturing requires V2X compliance for EU type-approval; US NHTSA V2X mandate could require V2X in new US Tesla vehicles; OTA-based rollout means V2X software can be added post-hardware-deployment	Roughly equal (both face similar regulatory timelines; Tesla’s EU manufacturing creates slightly more urgency)	NHTSA V2X mandate (if enacted 2026–2027) would require V2X in all new US vehicles, leveling the hardware baseline by 2030
Overall V2X verdict	V2X is currently a background technology for both Waymo and Tesla — neither depends on it, and both operate commercially without V2X integration. The 2026–2028 period will be decisive: NHTSA V2X rulemaking, EU ITS Delegated Act, and China’s C-V2X mandates will begin requiring V2X capability in new vehicles. Waymo’s constrained geofenced zones make V2I integration more tractable than Tesla’s nationwide geography, but Tesla’s fleet scale creates the largest theoretical V2V cooperative perception opportunity in the AV industry. The smart city integration trajectory points toward a hybrid architecture where camera/sensor-based AV perception is enhanced by V2X connectivity — but the timing and depth of V2X integration will be driven more by regulatory mandates than by either company’s voluntary investment.

Section 6 — About This Series

This is article 200 in the Physical AI Benchmark Series — a milestone article covering Vehicle-to-Everything (V2X) connectivity and smart city integration. Previous articles in this series have covered: Waymo’s commercial ramp index, the humanoid robot race, AV unit economics, global competitive dynamics, HD mapping technology, fleet operations, software and OTA updates, insurance and liability frameworks, consumer demand signals, partnerships and licensing, competitive moats, Cybercab versus Model Y robotaxi economics, AV safety data, Waymo Gen 6, Optimus manufacturing, physical AI scorecard snapshots, 2030 forecast scenarios, investor frameworks, Waymo’s city expansion pipeline, Tesla’s state approval map, AV weather and climate constraints, the physical AI talent war, the regulatory calendar, robotaxi fare pricing, the AV data flywheel comparison, humanoid deployment tracking, supply chain analysis, consumer adoption demand analysis, and V2X connectivity as today’s milestone topic.

The Physical AI Benchmark Series tracks the most important technology, regulatory, and commercial developments in autonomous vehicles, humanoid robotics, and AI-enabled physical systems. Articles 1–199 are available in the site archive. Article 200 marks the series at full V2X depth — the infrastructure integration layer that will define how autonomous vehicles participate in the cities of the 2030s.