Open Models Almanac
Methods

single mention · not yet filed

Blocking control that waits for each action to finish before predicting the next

Blocking control is a robot evaluation scheme in which each action is fully executed on the robot before the next one is predicted and executed.1 It controls system dynamics across methods with different latency, letting policy prediction quality be tested independently of prediction speed.2

Trade-off It removes the influence of varying inference speeds on task performance, at the cost of forcing faster methods to run slower.3

How this definition was written

Summarised by glm-5.3-flash from 1 source; every sentence is checked against the quote it cites.

  1. each action is fully executed on the robot before the next one is predicted by the policy and executed by the controller — OpenVLA: An Open-Source Vision-Language-Action Model
  2. This scheme controls system dynamics across methods with varying amounts of latency and thus allows us to test the quality of a policy’s action predictions, independent of its prediction speed — OpenVLA: An Open-Source Vision-Language-Action Model
  3. we evaluate with blocking control to remove the influence of varying inference speeds on task performance — OpenVLA: An Open-Source Vision-Language-Action Model

Also called blocking control.