Dominique Gruyer

Vision Enhancement in Homogeneous and Heterogeneous Fog

One source of accidents when driving a vehicle is the presence of fog. Fog fades the colors and reduces the contrasts in the scene with respect to their distances from the driver. Various camera-based Advanced Driver Assistance Systems (ADAS) can be improved if efficient algorithms are designed for visibility enhancement in road images. The visibility enhancement algorithm proposed in [1] is not optimized for road images. In this paper, we reformulate the problem as the inference of the local atmospheric veil from constraints. The algorithm in [1] thus becomes a particular case. From this new derivation, we propose to better handle road images by introducing an extra constraint taking into account that a large part of the image can be assumed to be a planar road. The advantages of the proposed local algorithm are the speed, the possibility to handle both color and gray-level images, and the small number of parameters. A new scheme is proposed for rating visibility enhancement algorithms based on the addition of several types of generated fog on synthetic and camera images. A comparative study and quantitative evaluation with other state-of-the-art algorithms is thus proposed. This evaluation demonstrates that the new algorithm produces better results with homogeneous fog and that it is able to deal better with the presence of heterogeneous fog. Finally, we also propose a model allowing to evaluate the potential safety benefit of an ADAS based on the display of defogged images.

Cooperative Fusion for Multi-Obstacles Detection With Use of Stereovision and Laser Scanner

We propose a new cooperative fusion approach between stereovision and laser scanner in order to take advantage of the best features and cope with the drawbacks of these two sensors to perform robust, accurate and real time-detection of multi-obstacles in the automotive context. The proposed system is able to estimate the position and the height, width and depth of generic obstacles at video frame rate (25 frames per second). The vehicle pitch, estimated by stereovision, is used to filter laser scanner raw data. Objects out of the road are removed using road lane information computed by stereovision. Various fusion schemes are proposed and one is experimented. Results of experiments in real driving situations (multi-pedestrians and multi-vehicles detection) are presented and stress the benefits of our approach.

Maneuver-Based Trajectory Planning for Highly Autonomous Vehicles on Real Road With Traffic and Driver Interaction

This paper presents the design and first test on a simulator of a vehicle trajectory-planning algorithm that adapts to traffic on a lane-structured infrastructure such as highways. The proposed algorithm is designed to run on a fail-safe embedded environment with low computational power, such as an engine control unit, to be implementable in commercial vehicles of the near future. The target platform has a clock frequency of less than 150 MHz, 150 kB RAM of memory, and a 3-MB program memory. The trajectory planning is performed by a two-step algorithm. The first step defines the feasible maneuvers with respect to the environment, aiming at minimizing the risk of a collision. The output of this step is a target group of maneuvers in the longitudinal direction (accelerating or decelerating), in the lateral direction (changing lanes), and in the combination of both directions. The second step is a more detailed evaluation of several possible trajectories within these maneuvers. The trajectories are optimized to additional performance indicators such as travel time, traffic rules, consumption, and comfort. The output of this module is a trajectory in the vehicle frame that represents the recommended vehicle state (position, heading, speed, and acceleration) for the following seconds.

Improved visibility of road scene images under heterogeneous fog

One source of accidents when driving a vehicle is the presence of homogeneous and heterogeneous fog. Fog fades the colors and reduces the contrast of the observed objects with respect to their distances. Various camera-based Advanced Driver Assistance Systems (ADAS) can be improved if efficient algorithms are designed for visibility enhancement of road images. The visibility enhancement algorithm proposed in [1] is not dedicated to road images and thus it leads to limited quality results on images of this kind. In this paper, we interpret the algorithm in [1] as the inference of the local atmospheric veil subject to two constraints. From this interpretation, we propose an extended algorithm which better handles road images by taking into account that a large part of the image can be assumed to be a planar road. The advantages of the proposed local algorithm are its speed, the possibility to handle both color images or gray-level images, and its small number of parameters. A comparative study and quantitative evaluation with other state-of-the-art algorithms is proposed on synthetic images with several types of generated fog. This evaluation demonstrates that the new algorithm produces similar quality results with homogeneous fog and that it is able to better deal with the presence of heterogeneous fog.

Highly Automated Driving on Highways Based on Legal Safety

This paper discusses driving system design based on traffic rules. This allows fully automated driving in an environment with human drivers, without necessarily changing equipment on other vehicles or infrastructure. It also facilitates cooperation between the driving system and the host driver during highly automated driving. The concept, referred to as legal safety, is illustrated for highly automated driving on highways with distance keeping, intelligent speed adaptation, and lane-changing functionalities. Requirements by legal safety on perception and control components are discussed. This paper presents the actual design of a legal safety decision component, which predicts object trajectories and calculates optimal subject trajectories. System implementation on automotive electronic control units and results on vehicle and simulator are discussed.

Data association with believe theory

In this paper, we present a method based on believe theory to combine expert opinion or symbolic sensor data. We consider applications with large frame of discernment and we propose generalisation for believe mass combination. In order to take into account of unknown hypothesis, we introduce a new framework for Dempsters combination: it is called the extended open world. This framework offers the possibility to have an opinion about the conflict between the experts and about the opportunity to introduce a new hypothesis in the frame of discernment. Some results highlight advantages of this framework in decision process.

Collision Probability Assessment for Speed Control

In order to navigate safely, it is important to detect and to react to a potentially dangerous situation. Such a situation can be underlined by a judicious use of the locations and the uncertainties of both the navigating vehicle and the obstacles. We propose to build an estimation of the collision probability from the environment perception with its probabilistic modelling. Then this probability is used for updating a braking order applied to our vehicle either to avoid or to mitigate a collision. The probability of collision is computed from a product of integrals of a product of Gaussians. The integrals take into account the uncertain configurations and the volume of both the vehicle and the obstacles.

Map-aided localization with lateral perception

Accurate localization of a vehicle is a challenging task as GPS available on the market are not designed for lane-level accuracy application. Although dead reckoning helps, cumulative errors from inertial sensors result in a integration drift. This paper presents a new method of localization based on sensors data fusion. An accurate digital map of the lane marking is used as a powerful additional sensor. Road markings are detected by processing two lateral cameras to estimate their distance to the vehicle. Coupled with the map data in a EKF filter it improves the ego-localization obtained with inertial and GPS measurements. The result is a vehicle localization at an ego-lane level of accuracy, with a lateral error of less than 10 centimeters.

Low cost IMU-Odometer-GPS ego localization for unusual maneuvers

This paper presents the problem of outdoor vehicle localization during unusual maneuvers with the Interacting Multiple Model (IMM) and Extended Kalman Filter (EKF) approaches. IMM, contrary to classical methods, is based on the discretization of the vehicle evolution space into simple maneuvers. Each maneuver is represented by a simple dynamic model such as a constant velocity or a constant turning model. This allows the method to be optimized for highly dynamic vehicles. In this work, we focus on unusual vehicle maneuvers during some special driving situations, including very strong accelerations, high speed turnings or backwards driving with stop stages. The presented results are based on real measurements collected from different scenarios. Based on an EKF vs. IMM comparison, these results show a real interest of using the IMM method in order to take into account unusual maneuvers.

A new multi-lanes detection using multi-camera for robust vehicle location

This paper deals with a new multi-lane markings detection and tracking system. The proposed system uses multiple cameras positioned differently in order to reduce different kind of perturbations, such as light sensitivity. The algorithm combines robust Kalman filtering and association based on belief theory to achieve multi-object tracking. Thus, the system provides the ability to track lane markings without any assumption on their number. It also proposes a new lane change management. To study this new system, the algorithm has been implemented on an embedded computer equipped with multiple cameras. We present experimental results obtained on a track. These results allow us to show important advantages of this new system and its robustness by comparing it to a classical system.