Contents

Update campaigns and in-vehicle orchestration

An update campaign selects which vehicles receive new software; an in-vehicle orchestrator coordinates installation, activation and recovery across their computers.

Why it exists

Delivering software to a vehicle is only part of an update. The vehicle must receive the right package, install it under suitable conditions and return to service with components that still work together.

A central computer might gain a new function that needs matching software in a zone controller and a brake electronic control unit (ECU). Updating only the central computer could leave incompatible interfaces. A lost cellular connection or an interrupted programming session must also have a defined outcome.

Campaign management controls deployment across the fleet. In-vehicle orchestration coordinates the work inside each vehicle. Together, they turn a software release into a controlled change.

What it is

An over-the-air (OTA) update campaign is a managed deployment to selected vehicles. It defines eligibility, software packages, rollout stages and conditions for continuing or stopping deployment.

The update orchestrator, sometimes called the update master, executes the vehicle’s update plan. It often runs on a central computer and coordinates local update agents, gateways and ECU bootloaders. Its location and responsibilities depend on the vehicle architecture.

The main responsibilities are:

Component Responsibility
Backend campaign service Select vehicles, release packages, monitor progress and control rollout.
Telematics control unit (TCU) Provide cellular connectivity and, depending on the design, download or cache packages.
In-vehicle orchestrator Check readiness, coordinate installation and activation, and track recovery.
Target ECU or local update agent Validate and install its software, then report its state.

An update plan describes more than files. It also identifies compatible hardware, required starting versions, dependencies, installation order and recovery actions.

How it works

From campaign to confirmed vehicle state
Select fleetBackend checks vehicle configuration
DownloadCellular delivery through the TCU
Check readinessOrchestrator validates packages and conditions
InstallDistribute to central, zone and endpoint computers
Activate and verifyStart a compatible software set
Report resultConfirm versions or enter recovery

Downloading, installing and activating are separate steps; each can have different conditions.

Select the vehicles. The backend matches vehicle identification numbers (VINs) against configuration data: hardware variants, installed software, optional equipment and other release restrictions. A VIN identifies a vehicle; it does not by itself describe everything fitted to it. The vehicle checks its actual configuration before accepting the plan.

The campaign starts with a limited group and expands in stages. Operators monitor download failures, installation results, recovery events and relevant post-update faults. Unexpected patterns can pause further rollout while engineers investigate.

Download the package. Data travels over cellular connectivity through the TCU. The vehicle may cache packages and resume interrupted transfers. Downloading can happen while the vehicle is in use if the design permits it; this does not imply that installation is allowed.

Before installation, update components verify package authenticity, integrity and applicability. An encrypted connection protects transport, but does not replace verification of the package itself.

Check preconditions. The orchestrator checks the conditions required for the particular update. These can include the vehicle being parked, propulsion being disabled, sufficient battery state and an agreed safe state. Available energy must cover programming, resets and recovery, including the low-voltage supply that powers controllers.

The vehicle also provides required user information and obtains consent where applicable. It explains expected downtime and unavailable functions. Conditions that can change need continued supervision or a defined response if they become invalid.

Distribute and install. The central computer sends software over the Ethernet backbone to zone controllers and other computers. A zone controller or gateway may forward programming traffic to endpoint ECUs over Controller Area Network with Flexible Data-rate (CAN FD).

Unified Diagnostic Services (UDS) commonly provides programming operations. Diagnostic communication over Internet Protocol (DoIP) can carry diagnostic traffic over Ethernet; a gateway may route it onward using the target network’s transport. DoIP does not itself define the fleet campaign or installation policy.

Where supported, software can be written to an inactive storage area in the background. Other ECUs must stop their application and enter a programming bootloader, so installation requires downtime. Background installation does not mean background activation: switching to the new software may still require a controlled restart.

Respect dependencies. The orchestrator follows a dependency plan rather than updating targets independently. For example, it may prepare new images across several controllers before allowing activation. If one target fails, it must prevent an unsupported combination from entering service.

A consistent software set is a tested combination of versions, configurations and interfaces. It does not require every ECU to share the same version number. Where simultaneous activation is impossible, the design needs supported intermediate combinations or a controlled period of unavailability.

Activate, verify and report. Activation may involve a restart, a storage-slot switch or another target-specific action. The vehicle confirms running versions, communication and required function checks before declaring success.

Results are stored locally and reported to the backend when connectivity is available. A completed download or successful transfer is not evidence that the new software is running correctly.

Handle failures. Interrupted downloads can often resume. Programming failures may require a retry, bootloader recovery, rollback to a previous image or workshop repair. Recovery depends on the target’s capabilities and the compatibility of the resulting software set. Pausing the fleet campaign must not abandon a vehicle already performing a critical programming step.

When to use it — and when not

Use campaign management for controlled fleet deployment. Use in-vehicle orchestration whenever an update needs coordinated conditions, dependencies or recovery across vehicle components.

A small update may need only a simple local installation plan. It still needs eligibility checks, trustworthy software and clear success criteria.

Use workshop programming when remote prerequisites cannot be met, recovery requires physical access or the target lacks a supported remote update path.

Best practices

  • Define compatible software sets and supported intermediate states before releasing packages.
  • Separate download, installation, activation and verification in both implementation and status reporting.
  • Check actual vehicle configuration and readiness locally.
  • Make interrupted work resumable where supported; preserve progress across resets.
  • Test failures at meaningful boundaries: transfer interruption, power loss, failed activation and missing targets.
  • Give users clear downtime information and give operators actionable failure reasons.

Common pitfalls

  • Selecting vehicles from stale backend inventory without checking their current configuration.
  • Treating successful ECU programming as successful vehicle activation.
  • Assuming every target supports rollback or background installation.
  • Updating dependent ECUs independently and leaving incompatible versions active.
  • Checking battery state once without managing energy throughout the operation.
  • Reporting a missing backend acknowledgement as an installation failure when the vehicle has simply lost connectivity.

What it means for your discipline

  • Software update & OTA Core

    Defines eligible vehicles, rollout stages, installation plans and recovery rules; monitors campaign results and pauses deployment when failures appear.

  • E/E architecture Partial

    Assigns orchestration responsibilities and defines compatible software sets across central computers, zone controllers and ECUs.

  • Diagnostics & lifecycle Partial

    Implements programming sessions, gateway routing and recovery access; makes ECU programming results useful for workshop troubleshooting.

  • Cybersecurity Partial

    Defines package authentication and update authorization; checks that recovery paths preserve software trust.

  • Functional safety Partial

    Defines safe installation conditions, activation checks and the vehicle behavior required when an update leaves a function unavailable.

No direct stake: Base software · HPC & application software · Hardware & silicon · Networks & communication

Where it sits in the software stack

Generic layered software of a central HPC (application processor with hypervisor, plus a safety island) and of a classic MCU-based ECU (AUTOSAR Classic). Highlighted modules matter for this topic. Click any module to open its page.

HPC · application processor (MPU) and safety island (MCU)
  • OTA client · UCM — Executes the vehicle update plan, checks readiness and dependencies, and coordinates installation, activation and recovery across targets.
  • Diagnostics — Provides diagnostic access to target versions, programming status and faults used to verify campaign results.
  • Security services — Verifies package authenticity and integrity and enforces authorization before installation.
MCU-based ECU · AUTOSAR Classic
  • Flash bootloader · secure boot — Receives and validates software, programs flash, and supports recovery after interrupted installation.
  • Diagnostics — Handles application-side programming requests, checks session conditions and reports faults before and after bootloader execution.