implement additional vType attributes (trac #2002)

[behrisch]

• maxSpeedLat (maximum lateral speed)
• restrict lateral movement in a single time step
• minGapLat (minimum lateral gap)
• should be added to width when keeping lateral distance to other vehicles

Migrated from http://sumo.dlr.de/ticket/2002

{
    "status": "closed", 
    "changetime": "2016-02-26T13:39:30Z", 
    "description": "* maxSpeedLat (maximum lateral speed)\n * restrict lateral movement in a single time step\n* minGapLat (minimum lateral gap)\n * should be added to width when keeping lateral distance to other vehicles\n\n", 
    "reporter": "namdre", 
    "cc": "", 
    "resolution": "fixed", 
    "_ts": "1456493970097458", 
    "component": "sharps", 
    "summary": "implement additional vType attributes", 
    "priority": "major", 
    "keywords": "", 
    "time": "2015-11-13T09:09:31Z", 
    "milestone": "0.26.0", 
    "owner": "", 
    "type": "task"
}

[behrisch]

@namdre committed dd6ea6a (aka r19621): {sharps} introducing new vType attributes refs #2002

[behrisch]

• @namdre changed _comment0 from:

minGapLat differs from minGap since two values of minGapLat are in effect between neighboring vehicles at the same time.

either a) Each vehicle only maintains its own minGap. In this case the vehicle with the lower minGapLat can ''push'' the other vehicle away b) Each vehicle takes the maximum value of both vehicles minGapLat (how would it know the other vehicles value?) c) The minGapLat between vehicles is the average value of both. This is easily implemented by computing a virtual vehicle width from width and minGapLat. Also, this is the thing that fits well with the sublane-based collision checking.

to:

1450258186915715

• @namdre changed _comment1 from:

minGapLat differs from minGap since two values of minGapLat are in effect between neighboring vehicles at the same time.

either a) Each vehicle only maintains its own minGap. In this case the vehicle with the lower minGapLat can ''push'' the other vehicle away b) Each vehicle takes the maximum value of both vehicles minGapLat (how would it know the other vehicles value?) c) The minGapLat between vehicles is the average value of both. This is easily implemented by computing a virtual vehicle width from width and minGapLat. Also, this is the thing that fits well with the sublane-based collision checking.

The issues is complicated by the fact that some vTypes such as bicycles require a larger gap and this is (should be) respected by all traffic participants.

to:

1450258382953661

• @namdre commented:

minGapLat differs from minGap since two values of minGapLat are in effect between neighboring vehicles at the same time.

either a) Each vehicle only maintains its own minGap. In this case the vehicle with the lower minGapLat can ''push'' the other vehicle away b) Each vehicle takes the maximum value of both vehicles minGapLat (how would it know the other vehicles value?) c) The minGapLat between vehicles is the average value of both. This is easily implemented by computing a virtual vehicle width from width and minGapLat. Also, this is the thing that fits well with the sublane-based collision checking.

The issues is complicated by the fact that some vTypes such as bicycles require a larger gap and this is (should be) respected by all traffic participants.

Furthermore, the lateral gaps observed in reality also depend on the speeds of the vehicles (relative and absolute).

[behrisch]

@namdre committed 41bb4a8 (aka r19624): {sharps} adding accessor refs #2002

[behrisch]

@namdre commented:

add a laneChangeModel parameter that computes minGap as alpha myMinGap + (1-alpha) otherMinGap

[behrisch]

@namdre committed 6806d55 (aka r20001): {sharps} now restricting lateral speed refs #2002

[behrisch]

@namdre committed ad123e6 (aka r20006): {sharps} added test for maxSpeedLat refs #2002

[behrisch]

@namdre committed f29bc3e (aka r20062): {sharps} added test for minGapLat refs #2002

[behrisch]

• @namdre commented:

work on lateralGap keeping continues in another ticket #2173

• @namdre changed resolution from "" to "fixed"
• @namdre changed status from "new" to "closed"