Valtech launches Nexus SDV platform for connected cars

Valtech has launched its Nexus SDV platform, built with Google Cloud Bigtable and Android Automotive OS software-defined vehicle components, for carmakers developing connected vehicle systems.

Nexus SDV is designed to connect vehicle software running on Android Automotive OS SDV with cloud-based data services, giving manufacturers a single framework for telemetry, service discovery, and software integration. It can also work with vehicle frameworks beyond Android Automotive OS through software development kits and a Synadia NATS interface.

The development reflects a broader shift in the car industry from hardware-led vehicle design to software-based architectures that separate digital services from individual control units. In practice, functions such as climate control, lighting, diagnostics, and remote monitoring can be managed as reusable software services rather than tied to specific electronic components.

Within the vehicle, the platform uses the Android Automotive OS SDV middleware layer to discover and manage available services and to stream high-frequency telemetry into Bigtable. According to Google, this allows services to continue operating independently of the main infotainment stack, including when a vehicle is parked and the primary screen system is powered down.

That setup is intended to support continuous remote monitoring without relying on the full infotainment environment. It also addresses a longstanding industry problem: software and telemetry systems have often been supplied by multiple vendors using separate pipelines and data formats, leaving carmakers with fragmented information.

Cloud backbone

Bigtable sits at the centre of the cloud side of the system. Google positions the database as well suited to large automotive telemetry workloads because it can ingest high volumes of time-series data while supporting low-latency access for analysis and downstream applications.

The database uses a sparse-row schema that can adapt as manufacturers add or change sensor inputs over time. That allows different data types, from engine measurements to more complex sensor outputs, to be stored in a unified table structure rather than split across multiple systems.

Continuous Materialized Views calculate metrics directly in the storage layer. Examples include average battery temperature and broader fleet measurements, which can then be used by software agents and analytical tools without repeating the same calculations elsewhere.

Bigtable also integrates with Google's Agent Development Kit and with Apache Spark-based workflows. In an automotive setting, this gives software agents access to live telemetry streams and historical fleet data, allowing them to trigger actions such as alerts, over-the-air software changes, or parts ordering when a specific pattern is detected.

Predictive use case

One of the main uses described for Nexus SDV is predictive maintenance. In this model, telemetry covering factors such as engine speed, vibration, fluid levels, and brake pressure is written into Bigtable, where live aggregates can be calculated and monitored for anomalies.

If an AI model identifies signs of wear or battery degradation, the system can assess the wider context, including mileage, service history, and planned journeys. It can then notify the driver through the in-car system, recommend a service booking at a nearby dealership, and begin ordering the required parts.

For manufacturers, the commercial goal is to cut warranty costs and reduce unexpected breakdowns by shifting maintenance decisions from fixed schedules to condition-based analysis. More broadly, Google and Valtech argue that a shared data layer can help carmakers build customer-facing services more quickly across the vehicle, mobile applications, and service networks.

Security model

Security features in the Nexus SDV architecture include mutual TLS, Google Cloud Certificate Authority Service, and private Google Kubernetes Engine clusters. Google said the design follows a defence-in-depth model intended to give vehicles secure identities and isolate network traffic.

The system also uses Google's Secure AI Framework to handle data privacy across machine learning processes. This is intended to protect both user data and manufacturers' intellectual property as more in-vehicle and fleet data is analysed by AI models.

Nexus SDV is available now as an open-source platform. Google said the relevant Android Automotive OS SDV support is part of the Android Automotive 26Q2 release, while Bigtable is already used in other automotive telemetry systems.

The companies are presenting the platform as a way for carmakers to avoid building the full software and data stack themselves. Instead, manufacturers can use a standardised base for vehicle connectivity and focus development on the services and interfaces that distinguish their brands.

The platform is optimised for Android Automotive OS SDV, but can also integrate with other vehicle frameworks.