Latent Terrain Representations for Trajectory Prediction. Feng, A. & Gordon, A. S. In Proceedings of the 1st ACM SIGSPATIAL International Workshop on Computing with Multifaceted Movement Data - MOVE'19, pages 1–4, Chicago, IL, USA, November, 2019. ACM Press. ![Latent Terrain Representations for Trajectory Prediction [link]](https://bibbase.org/img/filetypes/link.svg "link")
Paper doi abstract bibtex In natural outdoor environments, the shape of the surface terrain is an important factor in selecting a traversal path, both when operating off-road vehicles and maneuvering on foot. With the increased availability of digital elevation models for outdoor terrain, new opportunities exist to exploit this contextual information to improve automated path prediction. In this paper, we investigate predictive neural network models for outdoor trajectories that traverse terrain with known surface topography. We describe a method of encoding digital surface models as vectors in latent space using Wasserstein Autoencoders, and their use in convolutional neural networks that predict future trajectory positions from past trajectory data. We observe gains in predictive performance across three experiments, using both synthetic and recorded trajectories on real-world terrain.
@inproceedings{feng_latent_2019,
address = {Chicago, IL, USA},
title = {Latent {Terrain} {Representations} for {Trajectory} {Prediction}},
isbn = {978-1-4503-6951-0},
url = {http://dl.acm.org/citation.cfm?doid=3356392.3365218},
doi = {10.1145/3356392.3365218},
abstract = {In natural outdoor environments, the shape of the surface terrain is an important factor in selecting a traversal path, both when operating off-road vehicles and maneuvering on foot. With the increased availability of digital elevation models for outdoor terrain, new opportunities exist to exploit this contextual information to improve automated path prediction. In this paper, we investigate predictive neural network models for outdoor trajectories that traverse terrain with known surface topography. We describe a method of encoding digital surface models as vectors in latent space using Wasserstein Autoencoders, and their use in convolutional neural networks that predict future trajectory positions from past trajectory data. We observe gains in predictive performance across three experiments, using both synthetic and recorded trajectories on real-world terrain.},
booktitle = {Proceedings of the 1st {ACM} {SIGSPATIAL} {International} {Workshop} on {Computing} with {Multifaceted} {Movement} {Data} - {MOVE}'19},
publisher = {ACM Press},
author = {Feng, Andrew and Gordon, Andrew S.},
month = nov,
year = {2019},
keywords = {Narrative, STG, UARC},
pages = {1--4},
}
Downloads: 0
{"_id":"KHXqP8wQPygxaB3CB","bibbaseid":"feng-gordon-latentterrainrepresentationsfortrajectoryprediction-2019","author_short":["Feng, A.","Gordon, A. S."],"bibdata":{"bibtype":"inproceedings","type":"inproceedings","address":"Chicago, IL, USA","title":"Latent Terrain Representations for Trajectory Prediction","isbn":"978-1-4503-6951-0","url":"http://dl.acm.org/citation.cfm?doid=3356392.3365218","doi":"10.1145/3356392.3365218","abstract":"In natural outdoor environments, the shape of the surface terrain is an important factor in selecting a traversal path, both when operating off-road vehicles and maneuvering on foot. With the increased availability of digital elevation models for outdoor terrain, new opportunities exist to exploit this contextual information to improve automated path prediction. In this paper, we investigate predictive neural network models for outdoor trajectories that traverse terrain with known surface topography. We describe a method of encoding digital surface models as vectors in latent space using Wasserstein Autoencoders, and their use in convolutional neural networks that predict future trajectory positions from past trajectory data. We observe gains in predictive performance across three experiments, using both synthetic and recorded trajectories on real-world terrain.","booktitle":"Proceedings of the 1st ACM SIGSPATIAL International Workshop on Computing with Multifaceted Movement Data - MOVE'19","publisher":"ACM Press","author":[{"propositions":[],"lastnames":["Feng"],"firstnames":["Andrew"],"suffixes":[]},{"propositions":[],"lastnames":["Gordon"],"firstnames":["Andrew","S."],"suffixes":[]}],"month":"November","year":"2019","keywords":"Narrative, STG, UARC","pages":"1–4","bibtex":"@inproceedings{feng_latent_2019,\n\taddress = {Chicago, IL, USA},\n\ttitle = {Latent {Terrain} {Representations} for {Trajectory} {Prediction}},\n\tisbn = {978-1-4503-6951-0},\n\turl = {http://dl.acm.org/citation.cfm?doid=3356392.3365218},\n\tdoi = {10.1145/3356392.3365218},\n\tabstract = {In natural outdoor environments, the shape of the surface terrain is an important factor in selecting a traversal path, both when operating off-road vehicles and maneuvering on foot. With the increased availability of digital elevation models for outdoor terrain, new opportunities exist to exploit this contextual information to improve automated path prediction. In this paper, we investigate predictive neural network models for outdoor trajectories that traverse terrain with known surface topography. We describe a method of encoding digital surface models as vectors in latent space using Wasserstein Autoencoders, and their use in convolutional neural networks that predict future trajectory positions from past trajectory data. We observe gains in predictive performance across three experiments, using both synthetic and recorded trajectories on real-world terrain.},\n\tbooktitle = {Proceedings of the 1st {ACM} {SIGSPATIAL} {International} {Workshop} on {Computing} with {Multifaceted} {Movement} {Data} - {MOVE}'19},\n\tpublisher = {ACM Press},\n\tauthor = {Feng, Andrew and Gordon, Andrew S.},\n\tmonth = nov,\n\tyear = {2019},\n\tkeywords = {Narrative, STG, UARC},\n\tpages = {1--4},\n}\n\n","author_short":["Feng, A.","Gordon, A. S."],"key":"feng_latent_2019","id":"feng_latent_2019","bibbaseid":"feng-gordon-latentterrainrepresentationsfortrajectoryprediction-2019","role":"author","urls":{"Paper":"http://dl.acm.org/citation.cfm?doid=3356392.3365218"},"keyword":["Narrative","STG","UARC"],"metadata":{"authorlinks":{}},"html":""},"bibtype":"inproceedings","biburl":"https://api.zotero.org/users/6976806/collections/GUMH2QKL/items?key=ipCS99jY9KwOteQbpfAW7VKn&format=bibtex&limit=100","dataSources":["Z4B8L2qnYQgDdZhbe","jjKsXjebLR7xyJojc"],"keywords":["narrative","stg","uarc"],"search_terms":["latent","terrain","representations","trajectory","prediction","feng","gordon"],"title":"Latent Terrain Representations for Trajectory Prediction","year":2019}