Leveraging Scene Embeddings for Gradient-Based Motion Planning in Latent Space.
Jun YamadaChia-Man HungJack CollinsIoannis HavoutisIngmar PosnerPublished in: CoRR (2023)
Keyphrases
- motion planning
- latent space
- low dimensional
- latent variables
- degrees of freedom
- manifold learning
- gaussian process
- path planning
- dimensionality reduction
- mobile robot
- generative model
- parameter space
- gaussian processes
- lower dimensional
- humanoid robot
- feature space
- matrix factorization
- high dimensional
- video sequences
- gaussian process latent variable models
- multi robot
- probabilistic latent semantic analysis
- transfer learning
- distance metric
- three dimensional
- inverse kinematics
- gaussian mixture
- high dimensional data
- robotic arm
- image sequences
- prior knowledge
- probabilistic model
- moving objects
- collision free
- input image
- multi modal
- scene classification
- high dimensional spaces
- machine learning
- higher dimensional
- random projections
- feature extraction
- pattern recognition
- data points
- nearest neighbor
- bayesian framework
- vision system