SimuV4TaskList

Simuv4.0 task list

There are some new features already implemented in the new simulation. Have a look at it here.

Planned items for SD 2.1 are shortly described here, their effect on driving is also examined. When they have an effect on driving, the task should be implemented as an optional feature. Complex tasks will have their wiki page linked.

Bug fixes

Refine gear changing mechanism

Throttle and clutch fiddling is no longer wanted by the simulation engine, it will be done by the human or other robots completely (see [#735]). But it makes it possible to switch gears without any power loss, so to prevent "cheating", before the new gear is selected neutral will be used for a multiplier*{shift time} amount of time, with a multiplier <=1, its exact value needs to be tested.

Effect on driving: small, but this feature needs to be optional.

Fix tire friction at small speeds

Currently either the slip variable diverges, or when its cut to 0 friction also goes to 0. Resulting in wheel bouncing, cars sliding on banking, etc. Likely related bugs: [#743], [#744], [#199], [#177], [#620].

Effect on driving: marginal, but as it might affect how a car stops in the pit this feature needs to be optional until testing shows it is harmless.

New features

Torque calculation for force feedback

I suggest to put into tPrivCar, as tdble steeringwheelTq. A normalized torque is expected, being +1 when maximal turning torque to the right, -1 when to the left. First a linear centering force will be used, later something connected to tire friction will be added, based on the Pajecka-formula for torque.

Effect on driving: none for robots.

Turbocharged engine

Add a turbocharger, so that the engine torque depends not only on the throttle position and rev, but also on the turbochargers speed. Which in turn depends on the history of the engines torque output. A few new parameters will be required in the car.xml file, defaulting to values describing normally aspirated engines, that is what we have now.

Effect on driving: none for existing robots, as it will be added only for new or revised cars, which require new or revised robot setups anyway.

Redefine physical constants as parameter

For example gravitation or air density should be a parameter. This prepares us for changing weather and also for better track localisation (think Mexico City for example).

Effect on driving: hopefully none until the parameters have the same value as the constants now.

More Realistic Rev limiter

As described in [#781], the rev limiter should contain a level of hystersis so that the limit is far more noticable. This may affect the robot drivers slightly.