Modeling and Learning Behaviors
In order to safely navigate in a dynamic environment, a robot requires to know how the objects populating it will move in the future. Since this knowledge is seldom available, it is necessary to resort to motion prediction algorithms. Due to the difficulty of modeling the various factors that determine motion (e.g., internal state, perception), this is often done by applying machine-learning techniques to build a statistical model, using as input a collection of trajectories array through a sensor (e.g., camera, laser scanner), and then using that model to predict further motion.
This section describes the basic concepts involved in current motion learning and prediction approaches. After introducing the Bayes filter, it discusses Growing Hidden Markov Models, an approach which is able to perform lifelong learning, continuously updating its knowledge as more data are available. In experimental evaluation against two other state-of-the-art approaches, the presented approach consistently outperforms them regarding both prediction accuracy and model parsimony.
The section concludes with an overview of the current challenges and future research directions for motion modeling and learning algorithms.
KeywordsHide Markov Model Kalman Filter Discrete State Structure Learning Motion Prediction
- Brand M, Oliver N, Pentland A (1997) Coupled hidden markov models for complex action recognition. In: Proceedings of the 1997 conference on computer vision and pattern recognition, San Juan, pp 994–999Google Scholar
- Dee H-M (2005) Explaining visible behaviour. PhD thesis, University of LeedsGoogle Scholar
- Hoogs A, Perera AA (2008) Video activity recognition in the real world. In: Proceedings of the twenty-third AAAI conference on artificial intelligence, Chicago, pp 1551–1554Google Scholar
- Jockusch J, Ritter H (1999) An instantaneous topological map for correlated stimuli. In: Proceedings of the international joint conference on neural networks, vol 1, Washington DC, pp 529–534Google Scholar
- Liao L, Fox D, Kautz H (2004) Learning and inferring transportation routines. In: Proceedings of the national conference on artificial intelligence AAAI-04, AmsterdamGoogle Scholar
- Makris D, Ellis T (2002) Spatial and probabilistic modelling of pedestrian behavior. In: Proceedings of the British machine vision conference, Cardiff, pp 557–566Google Scholar
- Mozos OM (2008) Semantic place labeling with mobile robots. PhD thesis, University of Freiburg, FreiburgGoogle Scholar
- Neal RM, Hinton GE (1998) A new view of the em algorithm that justifies incremental, sparse and other variants. In: Jordan MI (ed) Learning in graphical models. Kluwer Academic, Dordrecht, pp 355–368Google Scholar
- Rabiner LR (1990) A tutorial on hidden markov models and selected applications in speech recognition. Read Speech Recog 77:267–296Google Scholar
- Reif J, Sharir M (1985) Motion planning in the presence of moving obstacles. In Symposium on the foundations of computer science, Portland, pp 144–154Google Scholar
- Vasquez D (2007) Incremental learning for motion prediction of pedestrians and vehicles. PhD thesis, Institut National Polytechnique de Grenoble, Grenoble. http://emotion.inrialpes.fr/bibemotion/2007/Vas07. Accessed 28 Sept 2011
- Walter M, Psarrow A, Gong S (1999) Learning prior and observation augmented density models for behaviour recognition. In: Proceedings of the British machine vision conference, Malvern, Worcestershire, pp 23–32Google Scholar