{"id":13755311,"url":"https://github.com/bayesian-object-tracking/dbrt","last_synced_at":"2025-05-10T01:30:35.019Z","repository":{"id":201483441,"uuid":"82816719","full_name":"bayesian-object-tracking/dbrt","owner":"bayesian-object-tracking","description":"Depth-Based Bayesian Robot Tracking","archived":false,"fork":false,"pushed_at":"2019-10-14T13:58:50.000Z","size":1612,"stargazers_count":20,"open_issues_count":2,"forks_count":7,"subscribers_count":8,"default_branch":"master","last_synced_at":"2024-11-16T09:33:58.807Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":"","language":"C++","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"gpl-3.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/bayesian-object-tracking.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}},"created_at":"2017-02-22T14:55:57.000Z","updated_at":"2023-05-09T03:17:47.000Z","dependencies_parsed_at":null,"dependency_job_id":"a836aa09-afff-4d5e-9971-c6adf1015d7e","html_url":"https://github.com/bayesian-object-tracking/dbrt","commit_stats":null,"previous_names":["bayesian-object-tracking/dbrt"],"tags_count":1,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/bayesian-object-tracking%2Fdbrt","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/bayesian-object-tracking%2Fdbrt/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/bayesian-object-tracking%2Fdbrt/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/bayesian-object-tracking%2Fdbrt/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/bayesian-object-tracking","download_url":"https://codeload.github.com/bayesian-object-tracking/dbrt/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":253349909,"owners_count":21894795,"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","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":[],"created_at":"2024-08-03T10:00:51.834Z","updated_at":"2025-05-10T01:30:34.641Z","avatar_url":"https://github.com/bayesian-object-tracking.png","language":"C++","funding_links":[],"categories":["C++"],"sub_categories":[],"readme":"# ROS Depth-Based Robot Tracking Package (dbrt)\n\nThis package extends the object tracking packages dbot and dbot_ros to track \narticulated rigid bodies with several degree of freedom. In addition to depth\nimages, the robot tracker incorporates joint angle measurements at a higher \nrate, typically 100Hz-1kHz. Here are some of the core features\n\n * Provides joint state estimates at the rate of joint encoders\n * Compensates for inaccurate kinematics by estimating biases on the joint \n   angles\n * Estimates the head camera to robot base, if needed. Typically, the exact \n   camera location is unknown\n * Handles occlusion\n * Copes with camera delays \n * Requires only the model, i.e. the URDF description including the link meshes.\n \nFor more details on the algorithm, please check https://am.is.tuebingen.mpg.de/publications/garciacifuentes-ral.\n\n## Getting Started Example\n\nFirst of all, set up and run the example, as described in the [Getting Started](https://github.com/bayesian-object-tracking/getting_started#robot-tracking)\ndocumentation.\n\n## Setting Up Your Own Robot\n\nNow you can use the working example as a starting point. To use your own robot, you will need\nits URDF, and you will need to modify some launch and config files in [dbrt_example](https://git-amd.tuebingen.mpg.de/open-source/dbrt_getting_started/tree/master/dbrt_example). The launch files\nshould be self explanatory and easy to adapt. You will need to edit \nthe file fusion_tracker_gpu.launch (fusion_tracker_cpu.launch) to use\nyour own robot model, instead of Apollo. \n\nThe main work will be to adapt the fusion_tracker_gpu.yaml \n(fusion_tracker_cpu.yaml) file to your robot. All the parameters \nfor the tracking algorithm are specified in this file, and it is robot\nspecific. You will have to adapt the link and joint names to your robot.\nFurthermore, you can specify which joints should be corrected using the \ndepth images, how aggressively they should be corrected, and whether\nyou want to estimate an offset between the true camera and the \nnominal camera in your robot model. \n\n### URDF Camera Frame\n\nOur algorithm assumes that the frame of the depth image (specified by\nthe camera_info topic) exists in your URDF robot model. You can check the camera frame\nby running:\n```bash\nrostopic echo /camera/depth/camera_info\n```\nIf this frame does not exist in your robot URDF, you have to add such a camera frame to the \npart of the robot where the camera is mounted. This requires \nconnecting a camera link through a joint to another link of the robot. Take a \nlook at [head.urdf.xacro](https://git-amd.tuebingen.mpg.de/open-source/dbrt_getting_started/blob/master/apollo_robot_model/models/head.urdf.xacro#L319) .\n\nThe XTION camera link *XTION_RGB* is connected to the link *B_HEAD* through the \njoint *XTION_JOINT*. The transformation between the camera and the robot is not \nrequired to be very precise, since our algorithm can estimate an offset. \nHowever, it must be accurate enough to provide \na rough initial pose.\n\n\n\n## How to cite?\n```\n@article{GarciaCifuentes.RAL,\n title = {Probabilistic Articulated Real-Time Tracking for Robot Manipulation},\n author = {Garcia Cifuentes, Cristina and Issac, Jan and W{\\\"u}thrich, Manuel and Schaal, Stefan and Bohg, Jeannette},\n journal = {IEEE Robotics and Automation Letters (RA-L)},\n volume = {2},\n number = {2},\n pages = {577-584},\n month = apr,\n year = {2017},\n month_numeric = {4}\n}\n```\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fbayesian-object-tracking%2Fdbrt","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fbayesian-object-tracking%2Fdbrt","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fbayesian-object-tracking%2Fdbrt/lists"}