扫地机器人I型墙测试数据

扫地机器人在工作时时常会遇到一些特殊场景，比如I型墙场景，基于视觉的扫地机在此场景中获取的深度数据常有边界不清晰、精度低等问题，如何能在此场景中获取精确数据以准确定位与避障是个难点也是重点，在此场景，扫地机深度视觉模组获取的深度数据的精度就显得十分重要。搭建扫地机真值系统，用真值相机和目标深度模组进行不同材质、光线、距离下I型墙数据采集，获取I形墙场景的数据，并对数据进行处理得到校正后的数据用于目标深度模组的校正，提升深度模组在扫地机I型墙场景下的测量精度，以解决扫地机定位与避障问题。采扫地机器人真值系统采集不同材质、光线、距离下的I型墙数据，包括原始深度相机数据、真值相机采集的点云数据；对原始深度相机数据进行系统误差去除；对真值相机采集的I型墙点云数据进行修复，通过平面拟合算法进行I型墙的墙面进行校正、再通过滤波算法对点云进行平滑去噪，得到修复后的真值相机点云；将修复后的真值相机点云与深度相机进行点云配准，得到校正后的I型墙的点云数据——校正后数据laser-x、laser-y、laser-z和校正后数据tof-x、tof-y、tof-z。

When working, robotic vacuum cleaners frequently encounter special scenarios such as the I-shaped wall scenario. For vision-based robotic vacuums, the depth data collected in this scenario often exhibits issues like blurry boundaries and low accuracy. Obtaining accurate depth data to enable precise positioning and obstacle avoidance is both a critical challenge and a key research focus, underscoring the importance of the accuracy of depth data acquired by the robotic vacuum's depth vision module in this scenario.

We developed a ground truth system for robotic vacuums, which collects I-shaped wall data under varying materials, lighting conditions and distances using a ground truth camera and the target depth module. We acquired datasets from the I-shaped wall scenario, and processed the collected data to generate calibrated data for calibrating the target depth module, so as to improve the measurement accuracy of the depth vision module in the I-shaped wall scenario of robotic vacuums, and address the positioning and obstacle avoidance problems of robotic vacuums.

The ground truth system collects I-shaped wall data under different materials, lighting conditions and distances, including raw depth camera data and point cloud data captured by the ground truth camera. First, systematic errors are removed from the raw depth camera data. Next, the I-shaped wall point cloud data collected by the ground truth camera is repaired: the wall surfaces of the I-shaped wall are calibrated via the plane fitting algorithm, followed by smoothing and denoising the point cloud through filtering algorithms, to obtain the repaired ground truth camera point cloud. Finally, the repaired ground truth camera point cloud is registered with the depth camera's point cloud to obtain calibrated I-shaped wall point cloud data, including calibrated data laser-x, laser-y, laser-z and calibrated data tof-x, tof-y, tof-z.

搜集汇总

数据集介绍

特点

扫地机器人I型墙测试数据包含1122条记录，涵盖多维度采集信息和校正后数据，旨在提升扫地机器人在I型墙场景下的深度测量精度，解决定位与避障问题。