Tekkaman

导航

 

Velocity Obstacle

  Two circular objects A,B, at time t(0), with velocity V(A),V(B). A represent the robot, and B represent obstacle.

    

  Collision Cone:

     

  V(A,B) is the relative velocity of A&B. V(A,B) = V(A) - V(B). And λ(A,B) is the line of V(A,B).

       

   By translating CC(A,B) by V(B),we get absolute Velocity Obstacle VO:

    

    

  The VO partitions the absolute velocities of A into avoiding & colliding velocities. Selecting V(A) outside of VO would avoid collision with B:

    

  Velocities on the boundaries of VO would result in A grazing B.

  To avoid multiple obstacles, we consider the union of the individual velocity obstacles:

    

  Since VO is based on a linear approximation of the obstacle's trajectory, using it to predict remote collisions may be inaccurate.
  To account for imminent collisions:

    

   

 【Avoidance Maneuver

  躲避策略。

  

  Feasible Acceleration:   

  RV: Reachable Velocity:

  

  

  Reachable Avoidance Velocity (RAV):

    

  

   A maneuver avoiding obstacle B can be computed by selecting any velocity in RAV.

Structure of the Avoidance Maneuvers

  躲避策略的结构。

posted on 2018-08-25 13:12  Tekkaman  阅读(660)  评论(0编辑  收藏  举报