本数据集用于记录线缆拖车-四足机器人系统在 L 形窄廊中的挑战性轨迹规划与真实跟踪结果。数据围绕宽度约 0.9 m 的窄廊实验场景，包含目标轨迹、真实运动轨迹、牵引机器人与拖车状态、线缆拉紧/松弛混合模式切换、速度和跟踪状态等信息。该算例利用混合动力学建模、前端混合搜索与后端非线性轨迹优化生成轨迹，并通过 Unitree A1 四足机器人与定制线缆拖车系统进行真实实验验证。数据可用于绘制目标轨迹、真实轨迹与跟踪状态曲线，评估混合模式轨迹规划在狭窄环境中的可行性、安全性和动态可执行性。
", "ds_source": "数据来源于论文《Hybrid Dynamics Modeling and Trajectory Planning for Cable-Trailer With Quadruped Robot System》的窄廊实验。本数据集展示了展示了 L 形窄廊中的目标轨迹、真实轨迹和跟踪状态；论文说明拖车通过线缆拉紧/松弛状态切换穿越 0.9 m 宽窄廊，搜索模块约用 1.46 ks 得到可行轨迹，优化模块约用 121.45 s 细化轨迹；仅保持拉紧模式的规划在搜索完全部节点后仍无法得到可行解。真实实验中系统平均速度约 0.76 m/s，最大线速度约 1.11 m/s，最大角速度约 0.26 rad/s，最终拖车位姿误差约 0.18 m 和 0.0024 rad。
", "ds_process_way": "基于 CT-QR 系统的平面运动状态构建混合动力学模型，将线缆状态区分为松弛模式和拉紧模式，并将拖车非完整约束、系统欠驱动特性以及线缆模式切换纳入前向动力学迭代。前端采用混合 A* 启发式搜索生成含模式切换的可行次优轨迹；后端以搜索轨迹为初值，利用 CasADi 建模并调用 IPOPT 求解非线性轨迹优化问题，综合考虑能耗、平滑性、动力学可行性和基于多边形几何的碰撞避免约束。真实实验数据通过对规划轨迹进行跟踪获得，并整理为目标轨迹、真实轨迹、跟踪状态和性能指标。
", "ds_quality": "本数据集为窄廊实验的规划与实物验证结果，包含仿真规划结果和真实系统跟踪结果。质量控制包括：采用混合动力学模型显式约束线缆拉紧/松弛状态、拖车非完整约束和输入/状态边界；采用几何多边形碰撞避免约束保证系统与障碍物之间的安全距离；使用混合搜索轨迹作为非线性优化初值以提高收敛性；通过真实 CT-QR 系统验证轨迹可执行性。系统能够在 0.9 m 宽 L 形通道中完成通过与停车，最终拖车位姿误差约为 0.18 m 和 0.0024 rad。
", "ds_acq_start_time": "2025-01-01 00:00:00", "ds_acq_end_time": null, "ds_acq_place": "实验室环境", "ds_acq_lon_east": null, "ds_acq_lat_south": null, "ds_acq_lon_west": null, "ds_acq_lat_north": null, "ds_acq_alt_low": null, "ds_acq_alt_high": null, "ds_share_type": "open-access", "ds_total_size": 30856, "ds_files_count": 0, "ds_format": "CSV", "ds_space_res": "", "ds_time_res": "", "ds_coordinate": "无", "ds_projection": "", "ds_thumbnail": "ee037b71-2055-4bc5-ad5d-87b9f8b73090.png", "ds_thumb_from": 0, "ds_ref_way": "", "paper_ref_way": "", "ds_ref_instruction": "本数据用于复现和分析论文 Fig. 8 中 L 形窄廊轨迹规划与真实跟踪结果，可用于比较混合模式规划与仅拉紧模式规划在狭窄环境中的可行性差异。使用时应说明该数据基于论文给定 CT-QR 系统参数、走廊几何、轨迹优化模型和实验平台，结论主要适用于该线缆拖车-四足机器人系统及相近的室内窄通道运动规划场景。", "ds_from_station": "", "organization_id": "9ecaaa78-39e9-411e-9f24-274e12aa643f", "ds_serv_man": "何心", "ds_serv_phone": "18961373056", "ds_serv_mail": "xinh@hust.edu.cn", "doi_value": "", "subject_codes": [ "410" ], "quality_level": 0, "publish_time": "2026-07-09 10:57:58", "last_updated": "2026-07-09 10:57:58", "protected": false, "protected_to": "2028-06-30 00:00:00", "lang": "zh", "cstr": "", "i18n": { "en": { "title": "Trajectory Planning Dataset for the Cable-Trailer Quadruped Robot in an L-Shaped Narrow Corridor", "ds_format": "CSV", "ds_source": "The data are derived from the narrow-corridor experiment in the paper \"Hybrid Dynamics Modeling and Trajectory Planning for Cable-Trailer With Quadruped Robot System.\" This dataset presents the target trajectory, real trajectory, and tracking states in an L-shaped corridor. The paper reports that the trailer passes through an approximately 0.9 m-wide corridor by exploiting taut/slack cable-mode transitions. The search module takes about 1.46 ks to find a feasible trajectory and the optimization module takes about 121.45 s to refine it; tension-only planning fails to find a feasible trajectory after all nodes are expanded. In the real experiment, the system reaches an average speed of about 0.76 m/s, a maximum linear speed of about 1.11 m/s, and a maximum angular speed of about 0.26 rad/s, with final trailer pose errors of about 0.18 m and 0.0024 rad.", "ds_quality": "The dataset consists of planning and hardware-validation results from the narrow-corridor experiment, including simulated planning outputs and real-system tracking data. Quality control includes explicit hybrid-dynamics constraints for taut/slack cable states, trailer nonholonomic constraints, and input/state bounds; geometric polygon collision-avoidance constraints to maintain clearance from obstacles; use of the hybrid-search trajectory as the initial guess for nonlinear optimization; and real CT-QR hardware validation. In the reported corridor experiment, the system passes through and stops in an approximately 0.9 m-wide L-shaped passage, with final trailer pose errors of about 0.18 m and 0.0024 rad. The paper does not disclose raw sensor-calibration errors or per-time-step noise statistics; therefore, the data are most suitable for method reproduction, figure regeneration, and algorithm comparison rather than generic statistical performance evaluation.", "ds_ref_way": "", "ds_abstract": "This dataset corresponds to Fig. 8 of the paper and records challenging trajectory planning and real-world tracking results for a cable-trailer with quadruped robot (CT-QR) system in an L-shaped narrow corridor. The data describe an approximately 0.9 m-wide corridor experiment and include the target trajectory, real motion trajectory, tractor and trailer states, taut/slack cable-mode transitions, velocity profiles, and tracking states. The trajectory is generated using hybrid dynamics modeling, front-end hybrid search, and back-end nonlinear trajectory optimization, and is validated on a Unitree A1 quadruped robot with a customized cable-trailer. The dataset can be used to reproduce the target trajectory, real trajectory, and tracking-state curves in Fig. 8, and to evaluate feasibility, safety, and dynamic executability of hybrid-mode trajectory planning in narrow environments.", "ds_time_res": "", "ds_acq_place": "Laboratory environment", "ds_space_res": "", "ds_projection": "", "ds_process_way": "The dataset is produced by constructing a planar hybrid dynamics model of the CT-QR system, where the cable state is divided into slack and taut modes and the trailer nonholonomic constraints, system underactuation, and cable-mode transitions are embedded in forward dynamics iterations. A front-end hybrid A* search generates a feasible suboptimal trajectory with mode transitions. The back-end uses the search trajectory as an initial guess, formulates the nonlinear trajectory optimization problem in CasADi, and solves it with IPOPT while considering energy, smoothness, dynamic feasibility, and polygon-based geometric collision avoidance constraints. Real-world data are obtained by tracking the planned trajectory and are organized into target trajectories, real trajectories, tracking states, and performance indices.", "ds_ref_instruction": "This dataset can be used to reproduce and analyze the L-shaped narrow-corridor trajectory planning and real-world tracking results shown in Fig. 8. It supports comparison between hybrid-mode planning and tension-only planning in narrow environments. Users should state that the data are based on the CT-QR system parameters, corridor geometry, trajectory optimization model, and experimental platform specified in the paper; the conclusions mainly apply to this cable-trailer quadruped robot system and similar indoor narrow-passage motion-planning scenarios." } }, "submit_center_id": "ncdc", "data_level": 0, "recommendation_value": 0, "license_type": "https://creativecommons.org/licenses/by/4.0/", "doi_reg_from": "reg_local", "cstr_reg_from": "reg_local", "doi_not_reg_reason": null, "cstr_not_reg_reason": null, "is_paper_in_submitting": false, "belong_to_nieer": false, "ds_topic_tags": [ "线缆拖车四足机器人", "混合动力学", "轨迹规划", "L形窄廊", "模式切换" ], "ds_subject_tags": [ "工程与技术科学基础学科" ], "ds_class_tags": [], "ds_locus_tags": [], "ds_time_tags": [], "ds_contributors": [ { "true_name": "张文涛", "email": "wentaozhang@hust.edu.cn", "work_for": "华中科技大学人工智能与自动化学院", "country": "中国" }, { "true_name": "徐少航", "email": "shaohangxu@hust.edu.cn", "work_for": "华中科技大学人工智能与自动化学院；香港城市大学数据科学系", "country": "中国" }, { "true_name": "左格为", "email": "gwzuo@hust.edu.cn", "work_for": "华中科技大学人工智能与自动化学院", "country": "中国" }, { "true_name": "李博林", "email": "bolin_li@hust.edu.cn", "work_for": "华中科技大学人工智能与自动化学院", "country": "中国" }, { "true_name": "王靖博", "email": "wangjingbo1219@gmail.com", "work_for": "上海人工智能实验室具身智能中心", "country": "中国" }, { "true_name": "朱立军", "email": "ljzhu@hust.edu.cn", "work_for": "华中科技大学人工智能与自动化学院", "country": "中国" } ], "ds_meta_authors": [ { "true_name": "朱立军", "email": "ljzhu@hust.edu.cn", "work_for": "华中科技大学人工智能与自动化学院", "country": "中国" }, { "true_name": "张文涛", "email": "wentaozhang@hust.edu.cn", "work_for": "华中科技大学人工智能与自动化学院", "country": "中国" } ], "ds_managers": [ { "true_name": "何心", "email": "xinh@hust.edu.cn", "work_for": "华中科技大学", "country": "中国" } ], "category": "其他" }