Monday, 15:15 - 15:40 h, Room: MA 313

 

Sebastian Albrecht
Inverse optimal control of human locomotion

Coauthors: Stefan Glasauer, Marion Leibold, Michael Ulbrich

 

Abstract:
The general hypothesis of our approach is that human motions are (approximately) optimal for an unknown cost function subject to the dynamics. Considering tasks where participants walk from a start to an end position and avoid collisions with crossing persons, the human dynamics are modeled macroscopically on a point-mass level. The locomotion problem results in an optimal control problem and in case of a crossing interferer an MPC-like approach seems suitable. The task of inverse optimal control is to find the cost function within a given parametrized family such that the solution of the corresponding optimal control problem approximates the recorded human data best. Our solution approach is based on a discretization of the continuous optimal control problem and on a reformulation of the bilevel problem by replacing the discretized optimal control problem by its KKT-conditions. The resulting mathematical program with complementarity conditions is solved by using a relaxation scheme and applying an interior-point solver. Numerical results for different navigation problems including hard and soft constraints in the optimal control problem are discussed.

 

Talk 1 of the invited session Mon.3.MA 313
"Optimization and equilibrium problems II" [...]
Cluster 3
"Complementarity & variational inequalities" [...]

 

  Payday Loans In Virginia. Of course, the choice is not that easy, as there exist great number of different preparations. Notwithstanding, Cialis is the one that definitely differs from all other products.