Mohammad Aldibaja

Improving localization accuracy for autonomous driving in snow-rain environments

Accurate localization is one of the most important issues for autonomous cars. This paper suggests a new LIDAR based method for improving the performance of localizing autonomous cars especially in snow-rain environment. Principal Component Analysis (PCA) is used to reconstruct LIDAR images in terms of enhancing quality and aligning pixel values to those in map images. In addition, edge-profile matching is incorporated to increase the accuracy of the lateral controlling in terms of reducing the effects of snow lines inside lanes. The real experimental results have verified that the proposed method is reliable and provides an acceptable localization error for driving autonomously in snow environments at maximum speed of 60 Km/h.

Simultaneous state recognition for multiple traffic signals on urban road

Automated vehicle researches move on to the public road experiments. This study focuses on a traffic signal detection based on mono-camera, predefined map database and accurate vehicle pose which is estimated by a localization module. By using the map data and the vehicle pose, an image ROI (Region-Of-Interest) can be calculated. This paper handles a situation with multiple signals appeared in an ROI especially for far traffic signals. We propose a probabilistic method for an image location of the target traffic signal in order to realize simultaneous traffic signal recognition for urban driving. In addition, an adaptive contrast updating strategy has been proposed to enhance the contrast of the extracted traffic signal using its probability. Experiments show the performance of the proposed method for real driving data.

LIDAR-data accumulation strategy to generate high definition maps for autonomous vehicles

Mapping is a very critical issue for enabling autonomous driving. This paper proposes a robust approach to generate high definition maps based on LIDAR point clouds and post-processed localization measurements. Many problems are addressed including quality, saving size, global labeling and processing time. High quality is guaranteed by accumulating and killing the sparsity of the point clouds in a very efficient way. The storing size is decreased using sub-image sampling of the entire map. The global labeling is achieved by continuously considering the top-left corner of the map images as a reference regardless to the driven distance and the vehicle orientation. The processing time is discussed in terms of using the generated maps in autonomous driving. Moreover, the paper highlights a method to increase the density of online LIDAR frames to be compatible with the intensity level of the generated maps. The proposed method was used since 2015 to generate maps of different areas and courses in Japan and USA with very high stability and accuracy.

Mono-camera based vehicle localization using lidar intensity map for automated driving

This paper reports an image-based localization for automated vehicle. The proposed method utilizes a mono-camera and an inertial measurement unit to estimate the vehicle pose. Self-localization is implemented by a map matching technique between the reference digital map and sensor observations. In general, the same types of sensors are used for map data and observations. However, this study is focused on the mono-camera based method using Lidar-based map for the purpose of a low-cost implementation. Image template matching is applied to provide a correlation distribution between the captured image and the predefined orthogonal map. A probability of the vehicle pose is then updated using the obtained correlation. The experiments were carried out for real driving data on an urban road. The results have verified that the proposed method estimates the vehicle position in 0.11[m] positioning errors on real-time.

Trajectory optimization and state selection for urban automated driving

The automated driving is an emerging technology in which a car performs recognition, decision making, and control. The decision-making system consists of route planning and trajectory planning. The route planning optimizes the shortest path to the destination like an automotive navigation system. According to static and dynamic obstacles around the vehicle, the trajectory planning generates lateral and longitudinal profiles for vehicle maneuver to drive the given path. This study is focused on the trajectory planning for vehicle maneuver in urban traffic scenes. This paper proposes a trajectory generation method that extends the existing method to generate more natural behavior with small acceleration and deceleration. This paper introduces an intermediate behavior to gradually switch from the velocity keeping to the distance keeping. The proposed method can generate smooth trajectory with small acceleration/deceleration. Numerical experiments show that the vehicle generates smooth behaviors according to surrounding vehicles.

Supervised Calibration Method for Improving Contrast and Intensity of LIDAR Laser Beams

Calibration of LIDAR laser beams in terms of contrast and intensity levels is very important for map generation and localization of autonomous vehicles. In this paper, we explain a simple semi-calibration method based on matching the shape and distribution of histograms. A laser beam output is selected to be a reference of the calibration process after manually tuning its intensity and contrast parameters to describe the road marks in prominent reflectivity. The histograms of the other laser beams are then aligned to the reference histogram and the calibration parameters of each beam are obtained. The experimental results have verified that the proposed method is reliable and provides a considerable enhancement of the map image quality as well as it improves the localization accuracy during the autonomous driving.