Adjusting timing of actuation commands to account for fixed and variable delays in autonomous driving control of vehicles

A method of controlling a vehicle in an autonomous driving mode, the method comprising:

• identifying, by one or more processors, a trajectory defining a future desired path for the vehicle to travel over a period of time;
• determining, by the one or more processors, a first delay value for a first portion of the period of time;
• generating, by the one or more processors, a first command to control the vehicle to take a first action when the first portion of the period of time begins, the first command identifying a first actuator of the vehicle;
• sending, by the one or more processors, the first command to the first actuator at a time based on the first delay value before the first portion of the period of time begins so that the first actuator causes the vehicle to move according to the first command during the first portion of the period of time;
• determining, by the one or more processors, a second delay value for a second portion of the period of time;
• generating, by the one or more processors, a second command to control the vehicle to take a second action when the second portion of the period of time begins, the second command identifying a second actuator of the vehicle;
• and
• sending, by the one or more processors, the second command to the second actuator at a time based on the second delay value before the second portion of the period of time begins so that the second actuator causes the vehicle to move according to the second command during the second portion of the period of time.