Pedestrian trajectory prediction in crossing scenario using fuzzy logic and switching Kalman filter
|
摘要
Pedestrian trajectory prediction plays an important role in bothadvanced driving assistance system(ADAS) and autonomous vehicles. An algorithm for pedestrian trajectory prediction in crossing scenario is proposed. To obtain features of pedestrian motion, we develop a method for data labelling and pedestrian body orientation regression. Using the hierarchical features as domain of discourse, fuzzy logic rules are built to describe the transition between different pedestrian states and motion models. With derived probability of each type of motion model we further predict the pedestrian trajectory in the next 1.5 s using switching Kalman filter(KF). The proposed algorithm is further verified in our dataset, and the result indicates that the proposed algorithm successfully predicts pedestrian's crossing behavior 0.4 s earlier before pedestrian moves. Meanwhile, the precision of predicted trajectory surpasses other methods including interacting multi-model KF and dynamic Bayesian network(DBN).
Pedestrian trajectory prediction plays an important role in bothadvanced driving assistance system(ADAS) and autonomous vehicles. An algorithm for pedestrian trajectory prediction in crossing scenario is proposed. To obtain features of pedestrian motion, we develop a method for data labelling and pedestrian body orientation regression. Using the hierarchical features as domain of discourse, fuzzy logic rules are built to describe the transition between different pedestrian states and motion models. With derived probability of each type of motion model we further predict the pedestrian trajectory in the next 1.5 s using switching Kalman filter(KF). The proposed algorithm is further verified in our dataset, and the result indicates that the proposed algorithm successfully predicts pedestrian's crossing behavior 0.4 s earlier before pedestrian moves. Meanwhile, the precision of predicted trajectory surpasses other methods including interacting multi-model KF and dynamic Bayesian network(DBN).
Pedestrian trajectory prediction plays an important role in bothadvanced driving assistance system(ADAS) and autonomous vehicles. An algorithm for pedestrian trajectory prediction in crossing scenario is proposed. To obtain features of pedestrian motion, we develop a method for data labelling and pedestrian body orientation regression. Using the hierarchical features as domain of discourse, fuzzy logic rules are built to describe the transition between different pedestrian states and motion models. With derived probability of each type of motion model we further predict the pedestrian trajectory in the next 1.5 s using switching Kalman filter(KF). The proposed algorithm is further verified in our dataset, and the result indicates that the proposed algorithm successfully predicts pedestrian's crossing behavior 0.4 s earlier before pedestrian moves. Meanwhile, the precision of predicted trajectory surpasses other methods including interacting multi-model KF and dynamic Bayesian network(DBN).
引文
1. Schneider N, Gavrila D M. Pedestrian path prediction with recursive Bayesian filters: a comparative study. German Conference on Pattern Recognition, 2013: 174-183
2. Goldhammer M, Gerhard M, Zernetsch S, et al. Early prediction of a pedestrian’s trajectory at intersections. International IEEE Conference on Intelligent Transportation Systems, IEEE, 2014: 237-242
3. Iryo-Asano M, Alhajyaseen W. Consideration of a pedestrian speed change model in the pedestrian-vehicle safety assessment of signalized crosswalks. Transportation Research Procedia, 2017, 21: 87-97
4. Kooij J F P, Schneider N, Flohr F, et al. Context-based pedestrian path prediction. European Conference on Computer Vision. Springer, Cham, 2014: 618-633
5. Karasev V, Ayvaci A, Heisele B, et al. Intent-aware long-term prediction of pedestrian motion. IEEE International Conference on Robotics and Automation, IEEE, 2016: 2543-2549
6. Asahara A, Maruyama K, Sato A, et al. Pedestrian-movement prediction based on mixed Markov-chain model. ACM Sigspatial International Symposium on Advances in Geographic Information Systems, ACM-GIS 2011, November 01-04, 2011, Chicago, Il, Usa, Proceedings, DBLP, 2011: 25-33
7. Rehder E, Kloeden H. Goal-directed pedestrian prediction. IEEE International Conference on Computer Vision Workshop, IEEE, 2015: 139-147
8. Kwak J Y, Lee E J, Ko B C, et al. Pedestrian’s intention prediction based on fuzzy finite automata and spatial-temporal features. Electronic Imaging, 2016, 2016(3): 1-6
9. Nasir M, Nahavandi S, Creighton D. Fuzzy simulation of pedestrian walking path considering local environmental stimuli. 2012, 19: 1-6
10. Nasir M, Lim C P, Nahavandi S, et al. A genetic fuzzy system to model pedestrian walking path in a built environment. Simulation Modelling Practice and Theory, 2014, 45(6): 18-34
11. Castro J L, Delgado M, Medina J, et al. An expert fuzzy system for predicting object collisions. Its application for avoiding pedestrian accidents. Expert Systems with Applications, 2011, 38(1): 486-494
12. Keller C G, Hermes C, Gavrila D M. Will the pedestrian cross? Probabilistic path prediction based on learned motion features. 2011
13. Keller C G, Gavrila D M. Will the pedestrian cross? A study on pedestrian path prediction. IEEE Transactions on Intelligent Transportation Systems, 2014, 15(2): 494-506
14. Quintero R, Almeida J, Llorca D F, et al. Pedestrian path prediction using body language traits. Intelligent Vehicles Symposium Proceedings, IEEE, 2014: 317-323
15. Chen Z, Ngai D C K, Yung N H C. Pedestrian behavior prediction based on motion patterns for vehicle-to-pedestrian collision avoidance. International IEEE Conference on Intelligent Transportation Systems, IEEE, 2008: 316-321
16. V?lz B, Mielenz H, Siegwart R, et al. Predicting pedestrian crossing using quantile regression forests. Intelligent Vehicles Symposium, IEEE, 2016: 426-432
17. Goldhammer M, K?hler S, Doll K, et al. Camera based pedestrian path prediction by means of polynomial least-squares approximation and multilayer perceptron neural networks. Sai Intelligent Systems Conference, IEEE, 2016: 390-399
18. Murphy K P. Switching Kalman filters, 1998
2. Goldhammer M, Gerhard M, Zernetsch S, et al. Early prediction of a pedestrian’s trajectory at intersections. International IEEE Conference on Intelligent Transportation Systems, IEEE, 2014: 237-242
3. Iryo-Asano M, Alhajyaseen W. Consideration of a pedestrian speed change model in the pedestrian-vehicle safety assessment of signalized crosswalks. Transportation Research Procedia, 2017, 21: 87-97
4. Kooij J F P, Schneider N, Flohr F, et al. Context-based pedestrian path prediction. European Conference on Computer Vision. Springer, Cham, 2014: 618-633
5. Karasev V, Ayvaci A, Heisele B, et al. Intent-aware long-term prediction of pedestrian motion. IEEE International Conference on Robotics and Automation, IEEE, 2016: 2543-2549
6. Asahara A, Maruyama K, Sato A, et al. Pedestrian-movement prediction based on mixed Markov-chain model. ACM Sigspatial International Symposium on Advances in Geographic Information Systems, ACM-GIS 2011, November 01-04, 2011, Chicago, Il, Usa, Proceedings, DBLP, 2011: 25-33
7. Rehder E, Kloeden H. Goal-directed pedestrian prediction. IEEE International Conference on Computer Vision Workshop, IEEE, 2015: 139-147
8. Kwak J Y, Lee E J, Ko B C, et al. Pedestrian’s intention prediction based on fuzzy finite automata and spatial-temporal features. Electronic Imaging, 2016, 2016(3): 1-6
9. Nasir M, Nahavandi S, Creighton D. Fuzzy simulation of pedestrian walking path considering local environmental stimuli. 2012, 19: 1-6
10. Nasir M, Lim C P, Nahavandi S, et al. A genetic fuzzy system to model pedestrian walking path in a built environment. Simulation Modelling Practice and Theory, 2014, 45(6): 18-34
11. Castro J L, Delgado M, Medina J, et al. An expert fuzzy system for predicting object collisions. Its application for avoiding pedestrian accidents. Expert Systems with Applications, 2011, 38(1): 486-494
12. Keller C G, Hermes C, Gavrila D M. Will the pedestrian cross? Probabilistic path prediction based on learned motion features. 2011
13. Keller C G, Gavrila D M. Will the pedestrian cross? A study on pedestrian path prediction. IEEE Transactions on Intelligent Transportation Systems, 2014, 15(2): 494-506
14. Quintero R, Almeida J, Llorca D F, et al. Pedestrian path prediction using body language traits. Intelligent Vehicles Symposium Proceedings, IEEE, 2014: 317-323
15. Chen Z, Ngai D C K, Yung N H C. Pedestrian behavior prediction based on motion patterns for vehicle-to-pedestrian collision avoidance. International IEEE Conference on Intelligent Transportation Systems, IEEE, 2008: 316-321
16. V?lz B, Mielenz H, Siegwart R, et al. Predicting pedestrian crossing using quantile regression forests. Intelligent Vehicles Symposium, IEEE, 2016: 426-432
17. Goldhammer M, K?hler S, Doll K, et al. Camera based pedestrian path prediction by means of polynomial least-squares approximation and multilayer perceptron neural networks. Sai Intelligent Systems Conference, IEEE, 2016: 390-399
18. Murphy K P. Switching Kalman filters, 1998