Inwook Shim

Urban structure classification using the 3D normal distribution transform for practical robot applications

Previous urban structure classification methods are intractable for practical robots in two viewpoints: storing point clouds and complex computation for using conditional random fields. This paper presents a classification method based on normal distribution transform (NDT) grids for practical robots. NDT grids store point clouds in the form of the mean and covariance, instead of directly dealing with huge point clouds. By taking the advantage of NDT grids, we design geometric-featured voxel (GFV) based on NDT grids to represent urban structures as a voxel model. The proposed method consists of three steps: GFV generation, segmentation, and classification. In the segmentation, GFVs are clustered according to units of urban structures by spectral clustering. For the classification, the clustered GFVs are classified as one kind of urban structures by supervised learning. Geometric characteristics of urban structures are expressed by a histogram of geometric words. Experimental results prove that the proposed method based on NDT grids is suitable for practical robots in terms of memory requirement, computation time, and even classification accuracy.

An Autonomous Driving System for Unknown Environments Using a Unified Map

Recently, there have been significant advances in self-driving cars, which will play key roles in future intelligent transportation systems. In order for these cars to be successfully deployed on real roads, they must be able to autonomously drive along collision-free paths while obeying traffic laws. In contrast to many existing approaches that use prebuilt maps of roads and traffic signals, we propose algorithms and systems using Unified Map built with various onboard sensors to detect obstacles, other cars, traffic signs, and pedestrians. The proposed map contains not only the information on real obstacles nearby but also traffic signs and pedestrians as virtual obstacles. Using this map, the path planner can efficiently find paths free from collisions while obeying traffic laws. The proposed algorithms were implemented on a commercial vehicle and successfully validated in various environments, including the 2012 Hyundai Autonomous Ground Vehicle Competition.

Robot System of DRC‐HUBO+ and Control Strategy of Team KAIST in DARPA Robotics Challenge Finals

This paper summarizes how Team KAIST prepared for the DARPA Robotics Challenge (DRC) Finals, especially in terms of the robot system and control strategy. To imitate the Fukushima nuclear disaster situation, the DRC performed a total of eight tasks and degraded communication conditions. This competition demanded various robotic technologies such as manipulation, mobility, telemetry, autonomy, localization, etc. Their systematic integration and the overall system robustness were also important issues in completing the challenge. In this sense, this paper presents a hardware and software system for the DRC-HUBO+, a humanoid robot that was used for the DRC; it also presents control methods such as inverse kinematics, compliance control, a walking algorithm, and a vision algorithm, all of which were implemented to accomplish the tasks. The strategies and operations for each task are briefly explained with vision algorithms. This paper summarizes what we learned from the DRC before the conclusion. In the competition, 25 international teams participated with their various robot platforms. We competed in this challenge using the DRC-HUBO+ and won first place in the competition.

Robotic software system for the disaster circumstances: System of team KAIST in the DARPA Robotics Challenge Finals

We developed a software system for operating the humanoid robot DRC-HUBO+ in disaster circumstances that the US Defense Advanced Research Projects Agency suggested. This system was developed under the consideration of the stability of whole system, the systemic software environment for multiple developers, and the recovery routine when the system encounters a seriously abnormal situation. With these design goals, we devised our strategy for three domains: system, vision, and communication. Following the strategy we made two software frameworks (PODO and VPC) and one logical flow of data for remote process management. With this software system, we conducted all Tasks in the DARPA Robotics Challenge Finals, and we won the competition.

Auto-adjusting camera exposure for outdoor robotics using gradient information

We present a new method to auto-adjust camera exposure for outdoor robotics. In outdoor environments, scene dynamic range may be wider than the dynamic range of the cameras due to sunlight and skylight. This can results in failures of vision-based algorithms because important image features are missing due to under-/over-saturation. To solve the problem, we adjust camera exposure to maximize image features in the gradient domain. By exploiting the gradient domain, our method naturally determines the proper exposure needed to capture important image features in a manner that is robust against illumination conditions. The proposed method is implemented using an off-the-shelf machine vision camera and is evaluated using outdoor robotics applications. Experimental results demonstrate the effectiveness of our method, which improves the performance of robot vision algorithms.

Geometrical Featured Voxel Based Urban Structure Recognition and 3-D Mapping for Unmanned Ground Vehicle

Recognition of structures in urban environments is a fundamental ability for unmanned ground vehicles. In this paper we propose the geometrical featured voxel which has not only 3-D coordinates but also the type of geometrical properties of point cloud. Instead of dealing with a huge amount of point cloud collected by range sensors in urban, the proposed voxel can efficiently represent and save 3-D urban structures without loss of geometrical properties. We also provide an urban structure classification algorithm by using the proposed voxel and machine learning techniques. The proposed method enables to recognize urban environments around unmanned ground vehicles quickly. In order to evaluate an ability of the proposed map representation and the urban structure classification algorithm, our vehicle equipped with the sensor system collected range data and pose data in campus and experimental results have been shown in this paper.

The Development of Sensor System and 3D World Modeling for Autonomous Vehicle

This paper describes a novel sensor system for 3D world modeling of an autonomous vehicle in large-scale outdoor environments. When an autonomous vehicle performs path planning and path following, well-constructed 3D world model of target environment is very important for analyze the environment and track the determined path. To generate well-construct 3D world model, we develop a novel sensor system. The proposed novel sensor system consists of two 2D laser scanners, two single cameras, a DGPS (Differential Global Positioning System) and an IMU (Inertial Measurement System). We verify the effectiveness of the proposed sensor system through experiment in large-scale outdoor environment.

Multi lidar system for fast obstacle detection

In recent year, much progress has been made in outdoor obstacle detection. However, for fast moving robotic platform, high-speed obstacle detection is still a daunting challenge. This paper describes laser based system for the fast obstacle detection. To do this, we introduce how to configure laser range finders by using a plane ruler for outdoor robotic platform. For high-speed obstacle detection, we use the gradient of points. We evaluate processing time and accuracy of our system by testing on real drive track including off-road course.

Vision system and depth processing for DRC-HUBO+

This paper presents a vision system and a depth processing algorithm for DRC-HUBO+, the winner of the DRC finals 2015. Our system is designed to reliably capture 3D information of a scene and objects and to be robust to challenging environment conditions. We also propose a depth-map upsampling method that produces an outliers-free depth map by explicitly handling depth outliers. Our system is suitable for robotic applications in which a robot interacts with the real-world, requiring accurate object detection and pose estimation. We evaluate our depth processing algorithm in comparison with state-of-the-art algorithms on several synthetic and real-world datasets.

Combinatorial approach for lane detection using image and LIDAR reflectance

Recently, lane detection algorithms have played significant roles in the field of vehicle technology. While many well working algorithms have been developed, they are hard to use in complex urban environments. In this paper, we propose an efficient approach for detecting lane markings using image information and LIDAR reflectance. The proposed algorithm has three phases: ground extraction, lane detections, and combining lane information. The proposed algorithm was implemented on a real vehicle and validated in various traffic environments, including the 2014 Hyundai Autonomous Vehicle Competition (AVC).