Skip to content

1. 26.07.0 Release

These release notes describe the changes introduced in Foretellix products in version 26.07.0. Features are classified as New, Changed, or Deprecated. "Changed" indicates either interface modifications or behavioral changes to existing functionality.

1.1 Release Highlights

  • Build compound intervals from Watchers & Checkers via the Python SDK — You can now programmatically create compound intervals from overlapping intervals in any Watcher/Checker combination (Watcher-Checker, Watcher-Watcher, Checker-Checker, Checker-Watcher) using the Foretify Manager Python SDK. This enables richer, scriptable coverage analysis.
  • More realistic VRU groups in generated scenarios — The generator now assigns uniformly distributed speeds across the members of a VRU (pedestrian/cyclist) group instead of one shared group speed. This produces more lifelike crowd and group-crossing behavior, improving the realism of safety-critical scenarios involving non-vehicle road users.
  • Contextual error messages in the Foretify Manager UI for VPlans — In the VPlan page, many generic "Fetch Request Error" messages are replaced with specific messages that state what failed and where. Users get actionable feedback pointing to the actual cause, cutting troubleshooting time and reducing the need to escalate to support.

1.1.1 Backward Compatibility

Compatibility Guidance for Foretellix Tool Versions

To ensure full support and optimal performance, we strongly recommend updating both Foretify and Foretify Manager to the same version. Backward compatibility between versions cannot always be guaranteed, although we are trying to keep it for two releases. When data is uploaded to Foretify Manager, it is automatically validated for compatibility.

Please note that forward compatibility—using results from newer versions of Foretify in older versions of Foretify Manager—is intentionally not supported.

Key Non-Compatible Changes in Recent Foretellix Releases

In Foretify Manager, issues have been removed from the VPlan page to simplify the UI and provide more clarity.

This change also removes the related Foretify Manager API surfaces. Any scripts or integrations using them must be updated:

  • REST endpoints removed:

    • GET /api/workspaces/{workspaceId}/failedRuns — failed runs for a VPlan section
    • POST /api/workspaces/{workspaceId}/runsBySection — runs by section and issue
    • POST /api/export/{workspaceId}/runIssuesToCSV — export run issues to CSV

  • Response fields removed:

    • runIssueCounters — on the workspace and every VPlan node (sections, structs, groups, items)
    • failedRuns — the per-issue counters on Test Run Groups (regressions) and on timeline points

  • Request/response types removed:

    • RunIssueCount
    • GetRunsBySectionAndIssueRequest
    • IssuesExportRequest

Please review all Features marked as [Changed] in this release for further details on backward compatibility.

1.2 Foretify

[New] Generation of VRU groups with uniformly distributed speeds (SDL-73580)

[New][Early Access] Added support for assigning individually varied speeds to pedestrians, cyclists, and VRU group members in OSC2 scenario code, enabling realistic crowd simulation where each actor moves at its own distinct speed.

  • Known limitations (Early Access): In this release, group member speeds can be drawn only from a uniform distribution (via speed_range) or assigned explicitly (via speeds); other built-in or user-defined distributions are not yet supported.
  • The following capabilities are planned for a future release and are not part of this Early Access:
    • Defining a multi-point VRU path as a single shaped movement (to avoid the generation slowdown of chaining individual moves)
    • Allowing collisions between VRUs within a group

For more information, refer to range type.

1.3 Foretify Manager

[New] Compare runs on the same map (SDL-76497)

  • You can now overlay any two runs that share the same map in the Single Run Debugger, regardless of whether one was produced by re-simulating the other. No Smart Replay ancestry is required, and comparing existing runs does not re-execute or re-simulate anything.
  • Start a comparison from the Runs table by selecting a run and clicking Compare runs, or from within the Single Run Debugger using the Compare data dropdown. The compared run is drawn in a distinct color.

For more information, refer to Compare runs on the same map.

[New] Scenario Designer adds support for synthetic static obstacles, e.g. traffic cones (SDL-73328)

[New] Create new compound intervals based on Watchers and Checkers via Python SDK (SDL-67233)

  • You can now create compound intervals via the Python SDK (UI support is not yet available) using the following interval type combinations:
    • Watcher-Watcher
    • Checker-Checker
    • Watcher-Checker
  • These combinations are in addition to the previously supported Matcher-Matcher and Matcher-Watcher or Matcher-Checker combinations.

For more details, see the relationship tables of compound intervals.

[Changed] Fixed Test Management's Environment Settings dropdown suggestions in Foretify Manager (SDL-77296)

  • Dropdowns now suggest only values previously used for that key, instead of pooling every value in the section (e.g. --map no longer suggests debug levels).

[Changed] Replace generic "Fetch Request Error" with contextual error messages in Foretify Manager UI — VPlan (SDL-74586)

Replaced the generic "Fetch Request Error" notification in the Foretify Manager VPlan with context-specific error messages. Each failing operation — saving workspace settings, deleting a test run, loading dashboard data, and others — now shows a notification that identifies what failed.

1.4 Evaluation Pipeline

[Changed] UI support for launching evaluations in Foretify Manager (SDL-76115)

You can now run drive data through the Evaluation Pipeline from the Foretify Manager UI instead of running it locally. Foretify Manager distributes the evaluation across multiple machines in parallel, typically in the cloud, for much faster processing of large data volumes than a single local machine.

For detailed information, see Running an evaluation in Foretify Manager

1.5 Known Limitations

Exception calculating workspace with many Compound rules (SDL-77259)

There is a limitation when calculating workspaces with many compound rules applied. We recommend that until this is fixed, limit the number of compound rules in a workspace to 100.

Including actors unsupported by the selected simulator configuration may result in errors in Scenario Designer (SDL-76699)

  • Scenario Designer may display a “Failed to fetch actor trajectories” error when attempting to include an actor for which no suitable model exists in the selected simulator configuration.
  • To work around this limitation, add actors individually and skip the unsupported actor.
  • Clearer UI indication for unsupported actors is planned.

Unsaved changes in Scenario Designer are lost when using the browser Back button (SDL-75521)

  • A future fix will add a warning to inform users about potential data loss.

The dispatcher/orchestrator currently only supports 1 replica (SDL-68883)

The Kubernetes dispatcher and orchestrator currently only support 1 replica.

Only one Orchestrator pair is supported per Foretify Manager. Each Orchestrator should use its own namespace.