{"id":31802075,"url":"https://github.com/cavendish518/le-nav","last_synced_at":"2025-10-11T00:16:42.478Z","repository":{"id":304788574,"uuid":"1015114079","full_name":"Cavendish518/LE-Nav","owner":"Cavendish518","description":"This work investigates automatic hyperparameter tuning for planners such as DWA and TEB, and our navigation framework LE-Nav can be used to adjust hyperparameters of any optimization-based planner.","archived":false,"fork":false,"pushed_at":"2025-07-15T10:48:54.000Z","size":19542,"stargazers_count":5,"open_issues_count":0,"forks_count":1,"subscribers_count":2,"default_branch":"main","last_synced_at":"2025-07-15T15:10:19.207Z","etag":null,"topics":["adaptive-learning","dwa","hyperparameter-tuning","llm","navigation","robotics","ros","teb","vla","vln"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/Cavendish518.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null}},"created_at":"2025-07-07T03:04:37.000Z","updated_at":"2025-07-15T10:53:21.000Z","dependencies_parsed_at":"2025-07-15T15:46:02.692Z","dependency_job_id":null,"html_url":"https://github.com/Cavendish518/LE-Nav","commit_stats":null,"previous_names":["cavendish518/le-nav"],"tags_count":null,"template":false,"template_full_name":null,"purl":"pkg:github/Cavendish518/LE-Nav","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Cavendish518%2FLE-Nav","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Cavendish518%2FLE-Nav/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Cavendish518%2FLE-Nav/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Cavendish518%2FLE-Nav/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/Cavendish518","download_url":"https://codeload.github.com/Cavendish518/LE-Nav/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Cavendish518%2FLE-Nav/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":279005657,"owners_count":26083941,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","status":"online","status_checked_at":"2025-10-10T02:00:06.843Z","response_time":62,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["adaptive-learning","dwa","hyperparameter-tuning","llm","navigation","robotics","ros","teb","vla","vln"],"created_at":"2025-10-11T00:16:41.131Z","updated_at":"2025-10-11T00:16:42.471Z","avatar_url":"https://github.com/Cavendish518.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Learning to Tune Like an Expert: Interpretable and Scene-Aware Navigation via MLLM Reasoning and CVAE-Based Adaptation\n\nThis repo is the official project repository of [\\[**LE-Nav**\\]](https://arxiv.org/pdf/2507.11001) ([\\[DEMO\\]](https://drive.google.com/file/d/1_XVsA-nbONcEre_OyEVM9BInMulYK7_r/view?usp=sharing)).\n\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"/fig/scene1.gif\" width=\"200\"/\u003e\n  \u003cimg src=\"/fig/scene2.gif\" width=\"200\"/\u003e\n  \u003cimg src=\"/fig/scene3.gif\" width=\"200\"/\u003e\n  \u003cimg src=\"/fig/scene4.gif\" width=\"200\"/\u003e\n\u003c/p\u003e\n\n\u003cdiv align=\"left\"\u003e\n\n## 1. Overview\n![image](fig/comparison.jpg)\n\nLE-Nav is an interpretable and adaptive navigation framework designed for service robots operating in dynamic, human-centric environments. Traditional navigation systems often struggle in such unstructured settings due to fixed parameters and poor generalization. LE-Nav addresses this by combining multi-modal large language models (MLLMs) with conditional variational autoencoders (CVAEs) for zero-shot scene understanding and expert-level parameter tuning.\n\n![image](fig/overview.jpg)\n\n## 2. Environment\nDownload the code and create environment.\n```\nconda env create -f environment.yml\n```\nYou can also try:\n```\nconda create --name readscene python=3.9\nconda activate readscene\n```\nInstall dependencies.\n```\npip install openai \nconda install pytorch==2.1.0 torchvision==0.16.0 torchaudio==2.1.0 pytorch-cuda=12.1 -c pytorch -c nvidia\npip install numpy==1.22.4\nconda install tensorboard\npip install ultralytics\n```\nor\n```\npip install -r requirements.txt\n```\n## 3. Training\nCollect the data for your planner. Customize your config.yaml.\n```\npython train_cvae.py\n```\nIn our case, we select the following eight key hyperparameters for TEB: max_vel_x, max_vel_theta, acc_lim_x, acc_lim_theta, weight_max_vel_x, weight_acc_lim_x, weight_acc_lim_theta, weight_optimaltime\nand eight key hyperparameters for DWA: max_vel_x, max_vel_theta, acc_lim_x, acc_lim_theta, path_distance_bias, goal_distance_bias, occdist_scale, forward_point_distance. In the case of DWA, when updating hyperparameter max_vel_x,\nwe additionally synchronize the value of max_vel_trans to be consistent with max_vel_x. During the deployment, the inflation_radius of global costmap is also recorded and learned as it is closely related to the local planner.\n\n## 4. Deployment\nFill in the path, api key in the ROS file. (Developed with ROS Melodic.)\n```\nsource ~/your_ws/devel/setup.bash\nrosrun your_package path/to/image_infer_node.py\n```\nIf you use the same setup, you can try the [\\[model parameters\\]](https://github.com/Cavendish518/LE-Nav/tree/main/weight) we provide.\n## 5. Citation\nIf your like our projects, please cite us and give this repo a star.\n```\n@article{wang2025learning,\n  title={Learning to Tune Like an Expert: Interpretable and Scene-Aware Navigation via MLLM Reasoning and CVAE-Based Adaptation},\n  author={Wang, Yanbo and Fang, Zipeng and Zhao, Lei and Chen, Weidong},\n  journal={arXiv preprint arXiv:2507.11001},\n  year={2025}\n}\n```\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fcavendish518%2Fle-nav","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fcavendish518%2Fle-nav","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fcavendish518%2Fle-nav/lists"}