Filters








47,265 Hits in 6.5 sec

Predicting Future Locations of Moving Objects by Recurrent Mixture Density Network

Rui Chen, Mingjian Chen, Wanli Li, Naikun Guo
2020 ISPRS International Journal of Geo-Information  
Accurate and timely location prediction of moving objects is crucial for intelligent transportation systems and traffic management.  ...  Motivated by the current study status, we are dedicated to a deep-learning-based approach to predict the coordinates of several future locations of moving objects based on recent trajectory records.  ...  Conflicts of Interest: The authors declare no conflicts of interest.  ... 
doi:10.3390/ijgi9020116 fatcat:ta6kqmolujfbfmna6svygebrku

Modelling pedestrian trajectory patterns with Gaussian processes

David Ellis, Eric Sommerlade, Ian Reid
2009 2009 IEEE 12th International Conference on Computer Vision Workshops, ICCV Workshops  
We illustrate the benefit of this approach for long term motion prediction where parametric models such as Kalman Filters would perform poorly.  ...  We propose a non-parametric model for pedestrian motion based on Gaussian Process regression, in which trajectory data are modelled by regressing relative motion against current position.  ...  Long term prediction can therefore be used to aid in target reacquisition.  ... 
doi:10.1109/iccvw.2009.5457470 dblp:conf/iccvw/EllisS009 fatcat:twq3a2i4c5ezrejd65rdbpymfi

Using accelerometer, high sample rate GPS and magnetometer data to develop a cattle movement and behaviour model

Y. Guo, G. Poulton, P. Corke, G.J. Bishop-Hurley, T. Wark, D.L. Swain
2009 Ecological Modelling  
For cows' movement between the "stay" areas a long-term prediction algorithm was implemented.  ...  Two learning algorithms were implemented: a Hidden Markov Model (HMM) and a long-term prediction-learning algorithm.  ...  The path by which the cow moved from one stay region to another was 18 generated using a long-term prediction learning process. 19 20 More details on the HMM design, the long-term prediction learning  ... 
doi:10.1016/j.ecolmodel.2009.04.047 fatcat:o44tm4lkgzdp7nsmjiuf5mnoda

A Mnemonic Kalman Filter for Non-Linear Systems with Extensive Temporal Dependencies

Steffen Jung, Isabel Schlangen, Alexander Charlish
2020 IEEE Signal Processing Letters  
Its true motion might depend on hundreds of parameters and can involve long-term temporal correlation.  ...  In particular, the Kalman Filter assumes prior and posterior Gaussian densities and is hence restricted to linear transition functions which are often insufficient to reflect the behaviour of a real object  ...  This Multivariate Density Long Short-Term Memory (MD-LSTM) network is then used in the KF prediction step to estimate the future state of a single object along with its covariance (Section III-B).  ... 
doi:10.1109/lsp.2020.3000679 fatcat:2hjpd7iwfza4pa5mnqmvrbrzsa

Motion prediction for moving objects: a statistical approach

D. Vasquez, T. Fraichard
2004 IEEE International Conference on Robotics and Automation, 2004. Proceedings. ICRA '04. 2004  
This paper proposes a technique to obtain long term estimates of the motion of a moving object in a structured environment.  ...  The results show that the technique is general, produces long-term predictions and is fast enough for its use in real time applications.  ...  Since they permit to take into account not only the current state of the object but also its past states, clusterbased techniques are by far the best ones when it comes to long term motion prediction.  ... 
doi:10.1109/robot.2004.1308883 dblp:conf/icra/VasquezF04 fatcat:hzmz4b6lkrhxlaeurjdq27xf3q

Predictive Navigation by Understanding Human Motion Patterns

Shu-Yun Chung, Han-Pang Huang
2011 International Journal of Advanced Robotic Systems  
To make robots coexist and share the environments with humans, robots should understand the behaviors or the intentions of humans and further predict their motions.  ...  Finally, the simulations and experiments are shown to validate the idea of this paper.  ...    Formulation of Prediction In prediction process, we split the problem into short term and long term prediction.  ... 
doi:10.5772/10529 fatcat:fo2jlr5w5fhebm3lkbkz63ekia

I-Planner: Intention-Aware Motion Planning Using Learning Based Human Motion Prediction [article]

Jae Sung Park and Chonhyon Park and Dinesh Manocha
2017 arXiv   pre-print
We demonstrate the benefits of our intention-aware planner in terms of computing safe trajectories in such uncertain environments.  ...  We represent the predicted human motion using a Gaussian distribution and compute tight upper bounds on collision probabilities for safe motion planning.  ...  The Gaussian Process regression algorithm corresponds to using Gaussian distribution ellipsoids around the predicted mean values of the joint positions.  ... 
arXiv:1608.04837v5 fatcat:64ygzlhu6fcrrjkqtlaf23xsju

Prediction-based Online Trajectory Compression [article]

Arlei Silva, Ramya Raghavendra, Mudhakar Srivatsa, Ambuj K. Singh
2016 arXiv   pre-print
ONTRAC learns prediction models for suppressing updates to a trajectory database using training data.  ...  Recent spatio-temporal data applications, such as car-shar\-ing and smart cities, impose new challenges regarding the scalability and timeliness of data processing systems.  ...  Scalability is measured in terms of the number of inserts processed per second and querying performance is evaluated in terms of the number of where (Definition 3) queries processed per second.  ... 
arXiv:1601.06316v2 fatcat:wpyhhypirbfqlhuzltzp55arw4

Vehicle Motion Prediction Algorithm with Driving Intention Classification

Wenda Ma, Zhihong Wu
2022 Applied Sciences  
based on the Gaussian process regression algorithm and horizontal heading angle prediction based on the long short-term memory method, which combines the predicted vehicle speed and heading angle to derive  ...  the future trajectory of the leading vehicle.  ...  In [10] , LSTM (long short-term memory) is used to learn the mapping from the input of the original sensor to the object coordinates in an unsupervised manner.  ... 
doi:10.3390/app12157443 fatcat:bb4plv5x4rdbjnkrexc2chgi4a

Online Long-Term Trajectory Prediction Based on Mined Route Patterns [chapter]

Petros Petrou, Panagiotis Tampakis, Harris Georgiou, Nikos Pelekis, Yannis Theodoridis
2020 Lecture Notes in Computer Science  
In this paper, we present a Big data framework for the prediction of streaming trajectory data by exploiting mined patterns of trajectories, allowing accurate long-term predictions with low latency.  ...  Subsequently, the trajectory prediction algorithm exploits these patterns in order to prolong the temporal horizon of useful predictions.  ...  This research has also been co-financed by the European Regional Development Fund of the European Union and Greek national funds through the Operational Program Competitiveness, Entrepreneurship and Innovation  ... 
doi:10.1007/978-3-030-38081-6_4 fatcat:6n5heoumwfguln3chpcp35uifm

Probabilistic collision estimation system for autonomous vehicles

Stefan Annell, Alexander Gratner, Lars Svensson
2016 2016 IEEE 19th International Conference on Intelligent Transportation Systems (ITSC)  
It is shown that in terms of robustness to noise the system successfully avoids collision.  ...  This system could either be used as information and feedback for a trajectory planner or as a support for decision making at higher level system monitoring.  ...  Long-term Trajectory Predictions The chosen approach for making long-term trajectory predictions in the system is based of the studies performed by [9] , [10] .  ... 
doi:10.1109/itsc.2016.7795597 dblp:conf/itsc/AnnellGS16 fatcat:n6cf6r77k5ckrd5zn2m5x2sxle

Approximate representations for multi-robot control policies that maximize mutual information

Benjamin Charrow, Vijay Kumar, Nathan Michael
2014 Autonomous Robots  
We address the problem of controlling a small team of robots to estimate the location of a mobile target using non-linear range-only sensors.  ...  using simulations and real world experiments in complex, indoor environments.  ...  Figure 6b shows the long term error from a separate experiment where the target moves 15 m to the right of its starting location and then back.  ... 
doi:10.1007/s10514-014-9411-2 fatcat:5py5uqbdv5hu5h44ql6do7l7rm

Approximate Representations for Multi-Robot Control Policies that Maximize Mutual Information

Benjamin Charrow, Vijay Kumar, Nathan Michael
2013 Robotics: Science and Systems IX  
We address the problem of controlling a small team of robots to estimate the location of a mobile target using non-linear range-only sensors.  ...  using simulations and real world experiments in complex, indoor environments.  ...  Figure 6b shows the long term error from a separate experiment where the target moves 15 m to the right of its starting location and then back.  ... 
doi:10.15607/rss.2013.ix.053 dblp:conf/rss/CharrowKM13 fatcat:d54mtgoiqveexix3gt4gds6nui

AutoTrajectory: Label-free Trajectory Extraction and Prediction from Videos using Dynamic Points [article]

Yuexin Ma, Xinge ZHU, Xinjing Cheng, Ruigang Yang, Jiming Liu, Dinesh Manocha
2020 arXiv   pre-print
To better capture the moving objects in videos, we introduce dynamic points.  ...  To the best of our knowledge, our method is the first to achieve unsupervised learning of trajectory extraction and prediction.  ...  Finally, we use these trajectories to train the trajectory prediction network. The whole process uses no labels.  ... 
arXiv:2007.05719v1 fatcat:kdpkhkezqrgazjeg5mwicyowwa

Model-Based Reinforcement Learning in Continuous Environments Using Real-Time Constrained Optimization

Olov Andersson, Fredrik Heintz, Patrick Doherty
2015 PROCEEDINGS OF THE THIRTIETH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE AND THE TWENTY-EIGHTH INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE  
This results in a bounded-rationality agent that makes decisions in real-time by efficiently solving a sequence of constrained optimization problems on learned sparse Gaussian process models.  ...  The efficacy of the approach is demonstrated on both an extended cart pole domain and a challenging quadcopter navigation task using real data.  ...  Combining model predictive control with Gaussian process models has also been previously suggested in the control community (Kocijan et al. 2004) where it was used in a small trajectory tracking problem  ... 
doi:10.1609/aaai.v29i1.9623 fatcat:mkm5ui5ymvd7hcwaq2iw52yrsa
« Previous Showing results 1 — 15 out of 47,265 results