{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/modular-vehicle-control-for-transferring","title":"Modular Vehicle Control for Transferring Semantic Information Between Weather Conditions Using GANs","arxiv_id":"1807.01001","date":"2018-07-03","proceeding":null,"authors":["Patrick Wenzel","Qadeer Khan","Daniel Cremers","Laura Leal-Taixé"],"abstract":"Even though end-to-end supervised learning has shown promising results for\nsensorimotor control of self-driving cars, its performance is greatly affected\nby the weather conditions under which it was trained, showing poor\ngeneralization to unseen conditions. In this paper, we show how knowledge can\nbe transferred using semantic maps to new weather conditions without the need\nto obtain new ground truth data. To this end, we propose to divide the task of\nvehicle control into two independent modules: a control module which is only\ntrained on one weather condition for which labeled steering data is available,\nand a perception module which is used as an interface between new weather\nconditions and the fixed control module. To generate the semantic data needed\nto train the perception module, we propose to use a generative adversarial\nnetwork (GAN)-based model to retrieve the semantic information for the new\nconditions in an unsupervised manner. We introduce a master-servant\narchitecture, where the master model (semantic labels available) trains the\nservant model (semantic labels not available). We show that our proposed method\ntrained with ground truth data for a single weather condition is capable of\nachieving similar results on the task of steering angle prediction as an\nend-to-end model trained with ground truth data of 15 different weather\nconditions.","url_abs":"http://arxiv.org/abs/1807.01001v2","url_pdf":"http://arxiv.org/pdf/1807.01001v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"modular-vehicle-control-for-transferring","repo_url":"https://github.com/pmwenzel/carla-domain-adaptation","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":null,"task_name":"Generative Adversarial Network"},{"task_slug":"self-driving-cars","task_name":"Self-Driving Cars"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}