TravelTo

We have implemented dribbling with 3 alternative ways: the options TravelTo(target), WalkTo(target, stopAtTarget) and WalkToTarget(speed, target).

WalkToTarget is the lower level state machine of the walking process, which fills the MotionRequest representation that will be passed to the Motion Control Module. Here, the ideal speed (for speedMode control) is limited when too close to an obstacle. This way, the robot inherently reduces the danger originated by such a condition.

WalkTo is a middle level state machine, it provides speed and intermediate target references to WalkToTarget to move the robot to a specific global target. By using the sonar and the foot contact sensors, we implement a later walk when in close proximity to other robots.

TravelTo is a high-level state machine. Using the robot detection network, it identifies free ranges and produces a sequence of global targets. These targets can be connected with an obstacle-free linear path, so can be directly provided to WalkTo as inputs.