Search

Radar sensors have great advantages over other sensors in estimating the motion states of moving objects, because they detect velocity components within one measurement cycle. Moreover, numerous successes have already been achieved regarding the classification of such objects. However, the advantage of instantaneous velocity measurement is lost when detecting static objects, so that their classification is much more demanding. In this paper, we use semantic segmentation networks to distinguish between frequently occurring infrastructure objects. The resulting semantic grids provide a location-based classification of the vehicle environment. Since even modern radars have a significantly poorer angular resolution than lidars, the relatively thin radar point cloud is accumulated in advance and transformed into 2D or 3D grids that act as network inputs. Occupancy grids are particularly advantageous here, since they calculate not only the obstacles but also the free spaces. With suitable parameter selection, which is very challenging due to the complexity of radar measurement, the resulting grids allow for good association with camera images. Finally, in order to evaluate possible advantages of 3D grids as network input with respect to the segmentation result, we created and evaluated a simulation dataset and two different real-world datasets in car parks and on motorways. As a result, Jaccard coefficients between 81% and 88% were achieved, depending on the dataset. It was also found that 3D input images lead to improvements in the car park dataset.

In the context of autonomous robots, one of the most important tasks is to prevent potential damage to the robot during navigation. For this purpose, it is often assumed that one must deal with known probabilistic obstacles, then compute the probability of collision with each obstacle. However, in complex scenarios or unstructured environments, it might be difficult to detect such obstacles. In these cases, a metric map is used, where each position stores the information of occupancy. The most common type of metric map is the Bayesian occupancy map. However, this type of map is not well suited for computing risk assessments for continuous paths due to its discrete nature. Hence, we introduce a novel type of map called the Lambda Field, which is specially designed for risk assessment. We first propose a way to compute such a map and the expectation of a generic risk over a path. Then, we demonstrate the benefits of our generic formulation with a use case defining the risk as the expected collision force over a path. Using this risk definition and the Lambda Field, we show that our framework is capable of doing classical path planning while having a physical-based metric. Furthermore, the Lambda Field gives a natural way to deal with unstructured environments, such as tall grass. Where standard environment representations would always generate trajectories going around such obstacles, our framework allows the robot to go through the grass while being aware of the risk taken., Published in the Special Issue "Frontiers in Mobile Robot Navigation" of Sensors. https://www.mdpi.com/1424-8220/21/22/7562

In this paper, two state-of-the-art solutions to the simultaneous localization and mapping (SLAM) problem are implemented, depending on the environment type. A line feature-based solution using the extended Kalman filter is selected for structured environments, while for unstructured an incremental likelihood maximization algorithm using scan matching is adopted. This work proposes an evaluation method for the mapping accuracy assessment, able to handle results from different map representations. The resulting maps of both algorithms are compared to the digitalized areas blueprints after being converted to a common representation, which makes use of custom elements supported by the popular OpenStreetMap digital map format. Experiments were performed in two parking garages with different characteristics showing the applicability of the proposed evaluation method independent of the SLAM algorithm used.

© 2017 IEEE. Robot simulations nowadays provide significant support in testing new algorithms for robotic systems for a broad area of tasks, including navigation, mapping, and SLAM. Within a simulation, special attention should be paid for providing algorithms with realistic testing environments that are to be further used for robot navigation. This paper presents an automatic tool that allows creating realistic landscapes in Gazebo simulation, which are based on results of real world sensor-based exploration. The tool provides automatic filtering and importing of an occupancy grid map into Gazebo framework as a heightmap.

International audience; Navigability assessment is useful for planning obstacles free trajectories in autonomous navigation tasks in 3D environments. Based on Geometric-Featured Voxel (GFV) maps developed in previous works, this paper proposes a method for 2D occupancy grid generation which is efficient for local navigation in natural environments. Built from a point cloud, a GFV map consists of voxels that are classified as tubular, planar, or scatter depending on the local spatial distribution of their inner points. Besides, each voxel is labelled as ground or non-ground considering slopes and rough surfaces found in unstructured scenarios. The proposed 2D occupancy grid represents accessible areas by considering the attributes and the height of each voxel over the ground level in relation to vehicle dimensions. The paper presents experimental results with 3D laser scans in natural environments.

© 2013 by MDPI (http://www.mdpi.org). Reproduction is permitted for noncommercial purposes., This paper presents a mapping system that is suitable for small mobile robots. An ad hoc algorithm for mapping based on the Occupancy Grid method has been developed. The algorithm includes some simplifications in order to be used with low-cost hardware resources. The proposed mapping system has been built in order to be completely autonomous and unassisted. The proposal has been tested with a mobile robot that uses infrared sensors to measure distances to obstacles and uses an ultrasonic beacon system for localization, besides wheel encoders. Finally, experimental results are presented., This work has been supported by the Spanish Ministerio de Ciencia e Innovación under project TEC2009-09871.

Poster; International audience; In this paper we have presented a fast algorithm to detect road borders from laser data. Two local search windows, one on right side of the host vehicle and the other on left, are moved right and left respectively from the current position of vehicle in map. A score function is evaluated to know the presence or absence of the road border in current search window. We have used the detected road border information to reduce false alarms in our previous work on DATMO (detection and tracking of moving objects). We also show how these information can be used to infer drivable area and the presence of intersections on the road. Results on data sets obtained from real demonstrator vehicles show that this technique can be successfully applied in real time.