Person Movement Prediction Using Neural Networks

Ubiquitous systems use context information to adapt appliance behavior to human needs. Even more convenience is reached if the appliance foresees the user’s desires and acts proactively. This paper proposes neural prediction techniques to anticipate a person’s next movement. We focus on neural predictors (multi-layer perceptron with back-propagation learning) with and without pre-training. The optimal configuration of the neural network is determined by evaluating movement sequences of real persons within an office building. The simulation results, obtained with one of the pre-trained neural predictors, show accuracy in next location prediction reaching up to 92%.

[1]  Lucian N. Vintan,et al.  Towards a high performance neural branch predictor , 1999, IJCNN'99. International Joint Conference on Neural Networks. Proceedings (Cat. No.99CH36339).

[2]  Rubén Anguera,et al.  Two-level branch prediction using neural networks , 2003, J. Syst. Archit..

[3]  Gordon B. Steven,et al.  Dynamic branch prediction using neural networks , 2001, Proceedings Euromicro Symposium on Digital Systems Design.

[4]  Michael C. Mozer,et al.  The Neural Network House: An Environment that Adapts to its Inhabitants , 1998 .

[5]  Wolfgang Trumler,et al.  Context Prediction Based on Branch Prediction Methods , 2003 .

[6]  Kristof Van Laerhoven,et al.  Real-time analysis of data from many sensors with neural networks , 2001, Proceedings Fifth International Symposium on Wearable Computers.

[7]  Wolfgang Trumler,et al.  Global State Context Prediction Techniques Applied to a Smart Office Building , 2004 .

[8]  Thomas G. Dietterich.  What is machine learning? , 2020, Archives of Disease in Childhood.

[9]  A POWERFUL,et al.  Romania towards a Powerful Dynamic Branch Predictor , 2003 .

[10]  Mario Aguilar,et al.  Prediction of pitch and yaw head movements via recurrent neural networks , 2003, Proceedings of the International Joint Conference on Neural Networks, 2003..