The emergence of data-driven approaches for control and planning in robotics have highlighted the need for developing experimental robotic platforms for data collection. However, their implementation is often complex and expensive, in particular for flying and terrestrial robots where the precise estimation of the position requires motion capture devices (MoCap) or Lidar. In order to simplify the use of a robotic platform dedicated to research on a wide range of indoor and outdoor environments, we present a data validation tool for ego-pose estimation that does not require any equipment other than the on-board camera. The method and tool allow a rapid, visual and quantitative evaluation of the quality of ego-pose sensors and are sensitive to different sources of flaws in the acquisition chain, ranging from desynchronization of the sensor flows to misevaluation of the geometric parameters of the robotic platform. Using computer vision, the information from the sensors is used to calculate the motion of a semantic scene point through its projection to the 2D image space of the on-board camera. The deviations of these keypoints from references created with a semi-automatic tool allow rapid and simple quality assessment of the data collected on the platform. To demonstrate the performance of our method, we evaluate it on two challenging standard UAV datasets as well as one dataset taken from a terrestrial robot.
MoCap2GT: A High-Precision Ground Truth Estimator for SLAM Benchmarking Based on Motion Capture and IMU Fusion
Shanghai Institute of Technical Physics, Chinese Academy of Sciences（上海技术物理研究所，中国科学院）；Advanced Display and Sensing Research Center, Yongjiang Laboratory（先进显示与传感研究中心，永江实验室）
Spatiotemporal Calibration and Ground Truth Estimation for High-Precision SLAM Benchmarking in Extended Reality
Advanced Display and Sensing Research Center, Yongjiang Laboratory, Ningbo, China（先进显示与感知研究中心，永江实验室，宁波，中国）；Shanghai Institute of Technical Physics, Chinese Academy of Sciences, Shanghai, China（上海技术物理研究所，中国科学院，上海，中国）
DiPE: Deeper into Photometric Errors for Unsupervised Learning of Depth and Ego-motion from Monocular Videos
School of Science and Engineering, The Chinese University of Hong Kong, Shenzhen（香港中文大学（深圳）理工学院）；Shenzhen Institute of Artificial Intelligence and Robotics for Society（深圳市人工智能与机器人研究院）
A Benchmark for Multi-Modal Lidar SLAM with Ground Truth in GNSS-Denied Environments
University of Turku（图尔库大学）
基于激光雷达的同时定位与地图构建（SLAM）方法在自主机器人系统中获得了相当大的成功。这在一定程度上是由于强大的SLAM算法的高精度以及新的低成本激光雷达产品的出现。这项研究对当前最先进的激光雷达SLAM算法进行了基准测试，其中多模态激光雷达传感器设置展示了不同的扫描模式（旋转和固态）和传感技术，以及安装在移动传感和计算平台上的激光雷达相机。我们扩展了我们以前的多模态多激光雷达数据集与额外的序列和新的来源的真值数据。具体而言，我们提出了一种新的多模态多激光雷达SLAM辅助和ICP为基础的传感器融合方法生成真值地图。有了这些地图，我们然后使用自然分布变换（NDT）方法匹配实时点云数据，以获得具有全6自由度位姿估计的真值。这种新颖的真值数据利用了高分辨率旋转和固态激光雷达。我们还包括新的开放道路序列与GNSS-RTK数据和额外的室内序列与运动捕捉（MOCAP）真值，补充以前的森林序列与MOCAP数据。我们进行了十个不同的SLAM算法和激光雷达组合实现的定位精度的分析。我们还报告了四种不同计算平台和总共五种设置（Intel和Jetson ARM CPU）的资源利用率。我们的实验结果表明，目前最先进的激光雷达SLAM算法执行不同类型的传感器非常不同。更多结果、代码和数据集可以在以下位置找到：\href{https://github.com/TIERS/tiers-lidars-datase…
Beyond Successful Robot Demonstrations: Learning from Human Interactions and Modeling Failure Dynamics
Zhejiang University（浙江大学）；Beijing Zhongguancun Academy（北京中关村学院）；University of Science and Technology of China（中国科学技术大学）；Eastern Institute of Technology（东方理工学院）；Institute of Automation, Chinese Academy of Sciences（中国科学院自动化研究所）
