Finding Optimal Robot Motion


Bryce Willey

November 16th, 2017
The Rice University logo
Dee Faught was built a robotic arm from Rice Students The NASA Mars Exploration Rover (Artists Concept)

Robots can...

  • assist people with disabilities/immobility

  • complete tasks that are unsafe or impossible for humans

Automating these tasks means allows widespread deployment

Motion planning is essential to automation.

Feasible Motion Planning

  • Finds a connected path from start to goal
  • Doesn't collide with any obstacles

Optimal Motion Planning

  • Assigns a cost to every path
  • Finds a feasible path that minimizes the cost

Pros

  • If a solution exists, it will be found
  • Very reliable
  • Can find optimal paths


Cons

  • Must smooth the path after finding it
  • Finding optimal paths is time consuming

Pros

  • Fast! (5-10x by our experiments)
  • Quality, smooth paths


Cons

  • No guarantee of finding a feasible path
  • Not as reliable as sampling planners

Currently, no rigorous comparisons in the literature

Optimization planners do special 'tricks' to speed up planning

What really makes optimization planners faster?

Abstract the 'tricks' from the planners


Test on a wider variety of planning problems


Use sampling and optimization together: quality motion and faster convergence

Sampling planners and optimization planners both have pros and cons

Optimization Planners are still not reliable enough to be used safely

Bonus Animations