Using basin ruins and co-moving low-dimensional latent coordinates for dynamic programming of biped walkers on roughing ground.
Hiromichi SuetaniAiko M. IdetaJun MorimotoPublished in: ICRA (2012)
Keyphrases
- low dimensional
- dynamic programming
- latent space
- high dimensional
- dimensionality reduction
- high dimensional data
- manifold learning
- humanoid robot
- principal component analysis
- control strategy
- data points
- latent variables
- greedy algorithm
- euclidean space
- optimal control
- stereo matching
- gaussian process latent variable models
- input space
- feature space
- linear dimensionality reduction
- ground surface
- river basin
- coarse to fine
- dp matching
- biped walking
- real time
- vector space
- linear programming
- control system
- lower dimensional
- multidimensional scaling
- control parameters
- nonlinear dimensionality reduction
- moving objects
- data sets