Unable to Implement RSS Distance Checking in Scenario Evaluation

[brkay54]

Checklist

• I've read the contribution guidelines.
• I've searched other issues and no duplicate issues were found.
• I've agreed with the maintainers that I can plan this task.

Background

The ODD Working Group has been discussing the implementation of RSS (Responsibility-Sensitive Safety) distance evaluation in running scenarios. The RSS distance, or minimum following distance, depends on dynamic variables such as the current velocity of the ego vehicle and the velocity of the vehicle it's following. However, these variables are not currently accessible in the scenario simulator, posing a challenge for planning evaluation.

Proposed Solutions

We've explored two potential approaches to address this issue:

Option 1: Enhance Scenario_Simulator_v2 to Publish Velocities

Description: Modify the existing Scenario_Simulator_v2 to publish the required velocity data.

Pros:

• Provides more accurate velocity values
• Enables precise RSS distance calculations

Cons:

• High implementation cost
• Requires significant changes to the current system

Status: Rejected due to resource constraints and implementation complexity.

Option 2: Integrate RSS Distance Metric into the Planning Evaluator

Description: Add an RSS distance metric calculation feature to the planning evaluator.

Pros:

• More cost-effective solution
• Easier to implement within the current framework
• Allows subscription to RSS distance in the scenario simulator

Cons:

• May be less accurate than real-time velocity data

Status: Currently the preferred option.

Request for Feedback

We welcome additional insights and opinions from the team. Please consider the following questions:

1. Are there any potential drawbacks to Option 2 that we haven't considered?
2. Do you foresee any integration challenges with the current planning evaluator?
3. Are there any alternative approaches we should explore?

Your input is valuable in ensuring we choose the most effective and efficient solution. Please share your thoughts and suggestions in the comments below.

[brkay54]

cc. @mitsudome-r @Igata-ctrl @kosuke55 @takayuki5168

[takayuki5168]

@brkay54
Both options sound good to me. I don't feel any drawbacks and challenges.

[stale]

This pull request has been automatically marked as stale because it has not had recent activity.