API review

Proposer: Jack O'Quin

Present at review:

• Piyush Khandelwal
• Corbyn Salisbury

This review mostly concerns the art_msgs/Observation and art_msgs/ObservationArray messages. Although defined in art_msgs, they describe the interface to art_observers.

Question / concerns / comments

Enter your thoughts on the API and any questions / concerns you have here. Please sign your name. Anything you want to address in the API review should be marked down here before the start of the meeting.

• (Jack) The art_msgs/Observation interface will be more convenient to use if we adopt the following conventions:

  • distance = std::numeric_limits<float>::infinity() when no object is detected.

    • (Piyush) +1

  • time = std::numeric_limits<float>::infinity() when no object is detected or the object getting farther away.

    • (Piyush) +1

• (Jack) The velocity field name is confusing, it's really a relative velocity. Rename the field accordingly.

  • set relative_velocity to std::numeric_limits<float>::quiet_NaN(), when no velocity estimate is available.

    • (Piyush) +1

• (Piyush) The clear field bothers me a bit - the way it is defined. For instance clear is set to true if the obstacle is as close as 1 meter away and be moving away slightly. Is this safe? (i.e. can we ever punch the gas and regret it). I suppose the vehicle following stuff works on this principle of the obstacle becoming constantly clear and not?

• (Piyush) What assumption is made when the applicable field is set to false?

• (Piyush) The clear field is currently redundant as it will always be automatically encoded by the time field.

Meeting agenda

To be filled out by proposer based on comments gathered during API review period

Conclusion

Package status change

• Action items that need to be taken.

• Major issues that need to be resolved

---