Evidential Grid Research Articles

A Dual Evidential Top-View Representation to Model the Semantic Environment of Automated Vehicles

We present a generic evidential grid mapping pipeline designed for imaging sensors such as LiDAR and cameras. Our grid-based evidential model contains semantic estimates for cell occupancy and ground separately. We specify the estimation steps for input data represented by point sets, but mainly focus on input data represented by images such as disparity maps or LiDAR range images. Instead of relying on an external ground segmentation only, we deduce occupancy evidence by analyzing the surface orientation around measurements. We conduct experiments and evaluate the presented method using LiDAR and stereo camera data recorded in real traffic scenarios. Our method estimates cell occupancy robustly and with a high level of detail while maximizing efficiency and minimizing the dependency to external processing modules.

Semantic Evidential Grid Mapping Using Monocular and Stereo Cameras.

Accurately estimating the current state of local traffic scenes is one of the key problems in the development of software components for automated vehicles. In addition to details on free space and drivability, static and dynamic traffic participants and information on the semantics may also be included in the desired representation. Multi-layer grid maps allow the inclusion of all of this information in a common representation. However, most existing grid mapping approaches only process range sensor measurements such as Lidar and Radar and solely model occupancy without semantic states. In order to add sensor redundancy and diversity, it is desired to add vision-based sensor setups in a common grid map representation. In this work, we present a semantic evidential grid mapping pipeline, including estimates for eight semantic classes, that is designed for straightforward fusion with range sensor data. Unlike other publications, our representation explicitly models uncertainties in the evidential model. We present results of our grid mapping pipeline based on a monocular vision setup and a stereo vision setup. Our mapping results are accurate and dense mapping due to the incorporation of a disparity- or depth-based ground surface estimation in the inverse perspective mapping. We conclude this paper by providing a detailed quantitative evaluation based on real traffic scenarios in the KITTI odometry benchmark dataset and demonstrating the advantages compared to other semantic grid mapping approaches.

Modeling evidential grids using semantic context information for dynamic scene perception

Uncertainty about urban environments stems not only from imprecise pose estimation and noisy information in images but also from the lack of semantic information. This article presents an approach to improve the perception capability of intelligent vehicles in complex urban environments. The new method uses the meta-knowledge extracted from semantic context images associated with depth information to model occupancy grids from stereo vision. It uses the evidential formalism of the Dempster–Shafer theory to manage uncertainties involved in grid discretization, partial observation of the environment and also dynamic elements present in the scene. Real experiments carried out in a challenging urban environment using the KITTI benchmark are reported, from which meaningful evaluations compared to the standard evidential grid are done to show that the proposed method is able to better handle semantic, dynamic and uncertainty aspects in the environment representation.

Fusion of neural networks, for LIDAR‐based evidential road mapping

AbstractLIDAR sensors are usually used to provide autonomous vehicles with three‐dimensional representations of their environment. In ideal conditions, geometrical models could detect the road in LIDAR scans, at the cost of a manual tuning of numerical constraints, and a lack of flexibility. We instead propose an evidential pipeline, to accumulate road detection results obtained from neural networks. First, we introduce RoadSeg, a new convolutional architecture that is optimized for road detection in LIDAR scans. RoadSeg is used to classify individual LIDAR points as either belonging to the road, or not. Yet, such point‐level classification results need to be converted into a dense representation, that can be used by an autonomous vehicle. We thus second present an evidential road mapping algorithm, that fuses consecutive road detection results. We benefitted from a reinterpretation of logistic classifiers, which can be seen as generating a collection of simple evidential mass functions. An evidential grid map that depicts the road can then be obtained, by projecting the classification results from RoadSeg into grid cells, and by handling moving objects via conflict analysis. The system was trained and evaluated on real‐life data. A python implementation maintains a 10 Hz framerate. Since road labels were needed for training, a soft labeling procedure, relying lane‐level HD maps, was used to generate coarse training and validation sets. An additional test set was manually labeled for evaluation purposes. So as to reach satisfactory results, the system fuses road detection results obtained from three variants of RoadSeg, processing different LIDAR features.

Evidential-Based Approach for Trajectory Planning With Tentacles, for Autonomous Vehicles

Driving is a combination of continuous mental risk assessment, sensory awareness, and judgment, all adapting to extremely variable surrounding conditions. Several research projects are working on autonomous vehicle to robotize this complex task. The work presented in this paper focuses on reactive local trajectory planning in uncertain environment. The environment uncertainty is one of the challenges that we face in trajectory planning. For autonomous vehicle, to be efficient, they need to be able to deal with this kind of uncertainty. In this paper, we show that the theory of belief functions with its ability to distinguish between different types of uncertainty is able to provide significant advantages in the context of trajectory planning. Using the belief functions, we build evidential grid that represents the surrounding environment. To plan a local trajectory, we generate a set of clothoid tentacles in the egocentred reference frame related to the ego-vehicle. Those tentacles represent possible trajectories that consider the current dynamical state of the vehicle and make a smooth variation in the vehicle dynamic variables. Once the representation of the environment and the possible trajectories are generated, an evaluation of each trajectory is carried out according to several criteria and the choice of the trajectory is made using the decision formalism of Markov decision process. To demonstrate the effectiveness of our evidential approach, we apply it to scenarios where ego-vehicle has to make decision in uncertain dynamical environments, using a driving simulator (SCANeR™ Studio).

A Study on Training Data Selection for Object Detection in Nighttime Traffic Scenes

Grid mapping is widely used to represent the environment surrounding a car or a robot for autonomous navigation. This paper describes an algorithm for evidential occupancy grid (OG) mapping that fuses measurements from different sensors, based on the Dempster-Shafer theory, and is intended for scenes with stationary and moving (dynamic) objects. Conventional OGmapping algorithms tend to struggle in the presence of moving objects because they do not explicitly distinguish between moving and stationary objects. In contrast, evidential OG mapping allows for dynamic and ambiguous states (e.g. a LIDAR measurement: cannot differentiate between moving and stationary objects) that are more aligned with measurements made by sensors. In this paper, we present a framework for fusing measurements as they are received from disparate sensors (e.g. radar, camera and LIDAR) using evidential grid mapping. With this approach, we can form a live map of the environment, and also alleviate the problem of having to synchronize sensors in time. We also designed a new inverse sensor model for radar that allows us to extract more information from object level measurements, by incorporating knowledge of the sensor’s characteristics. We have implemented our algorithm in the OpenVX framework to enable seamless integration into embedded platforms. Test results show compelling performance especially in the presence of moving objects.

Multi-Sensor Fusion in Dynamic Environment using Evidential Grid Mapping

The performance of autonomous agents in both commercial and consumer applications increases along with their situational awareness. Tasks such as obstacle avoidance, agent to agent interaction, and path planning are directly dependent upon their ability to convert sensor readings into scene understanding. Central to this is the ability to detect and recognize objects. Many object detection methodologies operate on a single modality such as vision or LiDAR. Camera-based object detection models benefit from an abundance of feature-rich information for classifying different types of objects. LiDAR-based object detection models use sparse point clouds, where each point contains accurate 3D position of object surfaces. Camera-based methods lack accurate object to lens distance measurements, while LiDAR-based methods lack dense feature-rich details. By utilizing information from both camera and LiDAR sensors, advanced object detection and identification is possible. In this work, we introduce a deep learning framework for fusing these modalities and produce a robust real-time 3D bounding box object detection network. We demonstrate qualitative and quantitative analysis of the proposed fusion model on the popular KITTI dataset.

Evidential SLAM Fusing 2D Laser Scanner and Stereo Camera

This work introduces a new complete Simultaneous Localization and Mapping (SLAM) framework that uses an enriched representation of the world based on sensor fusion and is able to simultaneously provide an accurate localization of the vehicle. A method to create an Evidential grid representation from two very different sensors, laser scanner and stereo camera, allows a better handling of the dynamic aspects of the urban environment and a proper management of errors to create a more reliable map, thus having a more precise localization. A life-long layer with high level states is presented, it maintains a global map of the entire vehicle’s trajectory and distinguishes between static and dynamic obstacles. Finally, we propose a method that at each current map creation estimates the vehicle’s position by a grid matching algorithm based on image registration techniques. Results on a real road dataset show that the environment mapping data can be improved by adding relevant information that could be missed without the proposed approach. Moreover, the proposed localization method is able to reduce the drift and improve the localization compared to other methods using similar configurations.

Grid-Based Environment Estimation Using Evidential Mapping and Particle Tracking

Modeling and estimating the current local environment by processing sensor measurement data is essential for intelligent vehicles. Static obstacles, dynamic objects, and free space have to be appropriately represented, classified, and filtered. Occupancy grids, known for mapping static environments, provide a common low-level representation using occupancy probabilities with an implicit data association through the discrete grid structure. Extending this idea toward dynamic environments with moving objects requires a static/dynamic classification of measured occupancy and a tracking of the dynamic state of grid cells. In this paper, we propose a new dynamic grid mapping approach. An evidential representation using the Dempster-Shafer framework is used to model hypotheses for static occupancy, dynamic occupancy, free space, and their combined hypotheses. These hypotheses are consistently estimated and accumulated in a dynamic grid map by an adapted evidential filtering, allowing one to distinguish static and dynamic occupancy. The evidential grid mapping is combined with a low-level particle filter tracking that is used to estimate cell velocity distributions and predict dynamic occupancy of the grid map. Static occupancy is directly modeled in the grid map without requiring particles, increasing efficiency, and improving the static/dynamic classification due to the persistent map accumulation. Experimental results with real sensor data show the effectiveness of the proposed approach in challenging scenarios with occlusions and dense traffic.

TRAJECTORY PLANNING FOR AUTONOMOUS VEHICLE IN UNCERTAIN ENVIRONMENT USING EVIDENTIAL GRID

This paper considers the trajectory planning problem for an autonomous vehicle given uncertain knowledge about the surrounding environment. We propose to use evidential occupancy grid to deal with sensor uncertainties. Our aim is to develop a planning approach based on clothoid tentacles allowing a vehicle to move autonomously and safely in an environment which is not perfectly known. First, we generate a set of clothoid tentacles in the egocentered reference frame related to the vehicle, then we evaluate each tentacle using several criteria including occupancy criterion with evidential grid.

Formal methods for reasoning and uncertainty reduction in evidential grid maps

Information fusion is the task of combining data collected from different sources into a unified representation. Here, a main challenge is to deal with the inherent uncertainty contained in the information, such as sensor noise, conflicting information, or incomplete knowledge. In current approaches, one usually employs independence assumptions in order to reduce the complexity. Because of this, the full potential of the gathered data is often not fully exploited and the fusion may lead to additional uncertainty. In order to reduce this uncertainty, further information in form of background and expert knowledge can be utilized, which is often available for real-world scenarios. However, reasoning on this knowledge is a computational complex task. In this work, we propose a methodology which utilizes formal methods for that reasoning, which allows to relax some of the independence assumptions. We demonstrate the proposed methodology using evidential grid maps – a belief function-based environment representation, in which different kinds of uncertainty are represented explicitly. Our methodology is evaluated based on basic structures as well as on real-world data sets. The results show that the uncertainty in the maps is significantly reduced by considering dependencies among cells.

An evidential approach to SLAM, path planning, and active exploration

Probability theory has become the standard framework in the field of mobile robotics because of the inherent uncertainty associated with sensing and acting. In this paper, we show that the theory of belief functions with its ability to distinguish between different types of uncertainty is able to provide significant advantages over probabilistic approaches in the context of robotics. We do so by presenting solutions to the essential problems of simultaneous localization and mapping (SLAM) and planning based on belief functions. For SLAM, we show how the joint belief function over the map and the robot's poses can be factored and efficiently approximated using a Rao-Blackwellized particle filter, resulting in a generalization of the popular probabilistic FastSLAM algorithm. Our SLAM algorithm produces occupancy grid maps where belief functions explicitly represent additional information about missing and conflicting measurements compared to probabilistic grid maps. The basis for this SLAM algorithm are forward and inverse sensor models, and we present general evidential models for range sensors like sonar and laser scanners. Using the generated evidential grid maps, we show how optimal decisions can be made for path planning and active exploration. To demonstrate the effectiveness of our evidential approach, we apply it to two real-world datasets where a mobile robot has to explore unknown environments and solve different planning problems. Finally, we provide a quantitative evaluation and show that the evidential approach outperforms a probabilistic one both in terms of map quality and navigation performance.

Map-Aided Evidential Grids for Driving Scene Understanding

Evidential grids have recently been shown to have interesting properties for mobile object perception. Possessing only partial information is a frequent situation when driving in complex urban areas, and by making use of the Dempster-Shafer framework, evidential grids are able to handle partial information efficiently. This article deals with a lidar perception scheme that is enhanced by geo-referenced maps used as an additional source of information in a multi-grid fusion framework. The paper looks at the key stages of such a data fusion process and presents an adaptation of the conjunctive combination rule for refining the analysis of conflicting information. This method relies on temporal accumulation to distinguish between stationary and moving objects, and applies contextual discounting for modeling information obsolescence. As a result, the method is able to better characterize the state of the occupied cells by differentiating moving objects, parked cars, urban infrastructure and buildings. Another advantage of this approach is its ability to separate the drivable from the non-drivable free space. Experiments carried out in real traffic conditions with a specially equipped car illustrate the performance of this approach.

Controlling Remanence in Evidential Grids Using Geodata for Dynamic Scene Perception

This article proposes a perception scheme in the field of intelligent vehicles. The method exploits prior map knowledge and makes use of evidential grids constructed from the sensor data. Evidential grids are based on occupancy grids and the formalism of the Dempster–Shafer theory. Prior knowledge is obtained from a geographic map which is considered as an additional source of information and combined with a grid representing sensor data. Since the vehicle environment is dynamic, stationary and mobile objects have to be distinguished. In order to achieve this objective, evidential conflict information is used for mobile cell detection. As well, an accumulator is introduced and used as a factor for mass function specialisation in order to detect static cells. data recorded in urban conditions illustrate the benefits of the presented approach.