The high diversity of urban environments, at both the inter and intra levels, poses challenges for robotics research. Such challenges include discrepancies in urban features between cities and the deterioration of sensor measurements within a city. With such diversity in consideration, this paper aims to provide Light Detection and Ranging (LiDAR) and image data acquired in complex urban environments. In contrast to existing datasets, the presented dataset encapsulates various complex urban features and addresses the major issues of complex urban areas, such as unreliable and sporadic Global Positioning System (GPS) data, multi-lane roads, complex building structures, and the abundance of highly dynamic objects. This paper provides two types of LiDAR sensor data (2D and 3D) as well as navigation sensor data with commercial-level accuracy and high-level accuracy. In addition, two levels of sensor data are provided for the purpose of assisting in the complete validation of algorithms using consumer-grade sensors. A forward-facing stereo camera was utilized to capture visual images of the environment and the position information of the vehicle that was estimated through simultaneous localization mapping (SLAM) are offered as a baseline. This paper presents 3D map data generated by the SLAM algorithm in the LASer (LAS) format for a wide array of research purposes, and a file player and a data viewer have been made available via the Github webpage to allow researchers to conveniently utilize the data in a Robot Operating System (ROS) environment. The provided file player is capable of sequentially publishing large quantities of data, similar to the rosbag player. The dataset in its entirety can be found at http://irap.kaist.ac.kr/dataset .

In this letter, we present a dataset capturing diverse visual data formats that target varying luminance conditions. While RGB cameras provide nourishing and intuitive information, changes in lighting conditions potentially result in catastrophic failure for robotic applications based on vision sensors. Approaches overcoming illumination problems have included developing more robust algorithms or other types of visual sensors, such as thermal and event cameras. Despite the alternative sensors’ potential, there still are few datasets with alternative vision sensors. Thus, we provided a dataset recorded from alternative vision sensors, by handheld or mounted on a car, repeatedly in the same space but in different conditions. We aim to acquire visible information from co-aligned alternative vision sensors. Our sensor system collects data more independently from visible light intensity by measuring the amount of infrared dissipation, depth by structured reflection, and instantaneous temporal changes in luminance. We provide these measurements along with inertial sensors and ground-truth for developing robust visual SLAM under poor illumination.

This paper presents a fast single image enhancement that is applicable regardless of channels in various environments. The main idea of the paper is combining model-based and fusion-based dehazing methods, thereby presenting balanced image enhancement while elaborating image details. The proposed method enhances both color and grayscale images without any prior information. Multiband decomposition is utilized to extract the base and detail layers for intensity and Laplacian modules. The proposed ambient map and transmission estimation for the intensity module are effective in restoring the true intensity. Adaptive nonlinear mapping functions adjust details on each residual layer. Through color-corrected reconstruction, our results demonstrate outstanding performance on various types of hazy images. The proposed method is thoroughly validated in terms of conventional image quality comparison. We also provide the evaluation at the application phase from both the semantic (segmentation) and geometric (direct odometry) vision based robotics application. The overall algorithm is presented in https://youtu.be/3Fk3kbaPkXQ.

We often encounter limited FOV situations due to various factors such as sensor fusion or sensor mount in real-world robot navigation. However, the limited FOV interrupts the generation of descriptions and impacts place recognition adversely. Therefore, we suffer from correcting accumulated drift errors in a consistent map using LiDAR-based place recognition with limited FOV. Thus, in this letter, we propose a robust LiDAR-based place recognition method for handling narrow FOV scenarios. The proposed method establishes spatial organization based on the range-elevation bin and azimuth-elevation bin to represent places. In addition, we achieve a robust place description through reweighting based on vertical direction information. Based on these representations, our method enables addressing rotational changes and determining the initial heading. Additionally, we designed a lightweight and fast approach for the robot's onboard autonomy. For rigorous validation, the proposed method was tested across various LiDAR place recognition scenarios (i.e., single-session, multi-session, and multi-robot scenarios). To the best of our knowledge, we report the first method to cope with the restricted FOV.

Illumination variance is a major challenge for vision-based robotics. Most approaches focus on alleviating illumination changes in already captured images. Despite the large utility, camera attributes have been empirically determined to function in a highly passive manner, yielding vision algorithm failure under radical illumination variance. Recent studies have proposed exposure and gain control schemes that could maximize image information and eschew saturation. In this article, we propose a proactive control scheme for the camera's two dominant attributes—exposure time and gain control. Unlike existing approaches, we formulate this camera attribute control as an optimization problem in which the underlying function is not known a priori. We first define a new metric of the image regarding these two major attributes to include both image gradients and signal-to-noise ratio simultaneously. Based on this metric, we introduce a new formulation for this attribute control via Bayesian optimization (BO) and learn the environmental change from the captured image. During the control, to mitigate the burden of image acquisition and Bayesian optimization, images are synthesized using a camera response function and avoided the actual frame grab from the camera. The proposed method was validated in light-flickering indoor, outdoor near sunset, and indoor–outdoor transient environments where light changes rapidly, supporting 20–40 Hz frame rates.

In this letter, we propose a method for accurately estimating the 6-Degree Of Freedom (DOF) pose in an urban environment when a High Definition (HD) map is available. An HD map expresses 3D geometric data with semantic information in a compressed format and thus is more memory-efficient than point cloud maps. The small capacity of HD maps can be a significant advantage for autonomous vehicles in terms of map storage and updates within a large urban area. Unfortunately, existing approaches failed to sufficiently exploit HD maps by only estimating partial pose. In this study, we present a full 6-DOF localization against an HD map using an onboard stereo camera with semantic information from roads. We introduce an 8-bit representation for road information, which allow for effective bitwise operation when matching between query data and the HD map. For the pose estimation, we leverage a particle filter followed by a full 6-DOF pose optimization. Our experimental results show a median error of approximately 0.3 m in the lateral and longitudinal directions for a drive of approximately 11 km. These results can be used by autonomous vehicles to correct the global position without using Global Positioning System (GPS) data in highly complex urban environments. The median operation speed is approximately 60 msec supporting 10 Hz.