Matthias Rapp

Automotive radar gridmap representations

In robotic applications gridmaps are a common representation of the environment. For the automotive field, radar as sensing technology is suitable due to its robustness. This paper presents two radar-based grid-mapping algorithms for automotive applications like self-localization. These algorithms involve first an amplitude-based approach, which gains information about the RCS of all targets, and second an occupancy grid-mapping approach with an adapted inverse sensor measurement model. Experiments show that both gridmapping algorithms result in adequate representations of the environment.

Semi-Markov Process Based Localization Using Radar in Dynamic Environments

Automotive localization in urban environment faces natural long-term changes of the surroundings. In this work, a robust Monte-Carlo based localization is presented. Robustness is achieved through a stochastic analysis of previous observations of the area of interest. The model uses a grid-based Markov chain to instantly model changes. An extension of this model by a Lévy process allows statements about reliability and prediction for each cell of the grid. Experiments with a vehicle equipped with four short range radars show the localization accuracy performance improvement in a dynamic environment.

A Feature-Based Approach for Group-Wise Grid Map Registration

For autonomous vehicles and advanced driver assistance systems, information on the actual state of the environment is fundamental for localization and mapping tasks. Localization benefits from multiple observations of the same location at different times as these may provide important information on static and mobile objects. For efficient mapping, the environment may be explored in parallel. For these purposes, multiple observations represented by grid maps have to be aligned into one mutual frame. This paper addresses the problem of group-wise grid map registration using an image processing approach. For registration, a rotational-invariant descriptor is proposed in order to provide the correspondences of points of interest in radar-based occupancy grid maps. As pairwise registration of multiple grid maps suffers from bias, this paper proposes a graph-based approach for robust registration of multiple grid maps. This will facilitate highly accurate range sensor maps for the aforementioned purposes. Large-scale experiments show the benefit of the proposed methods and compare it to state-of-the-art algorithms on radar measurements.

A fast probabilistic ego-motion estimation framework for radar

This paper presents a fast, joint spatial- and Doppler velocity-based, probabilistic approach for ego-motion estimation for single and multiple radar-equipped robots. The normal distribution transform is used for the fast and accurate position matching of consecutive radar detections. This registration technique is successfully applied to laser-based scan matching. To overcome discontinuities of the original normal distribution approach, an appropriate clustering technique provides a globally smooth mixed-Gaussian representation. It is shown how this matching approach can be significantly improved by taking the Doppler information into account. The Doppler information is used in a density-based approach to extend the position matching to a joint likelihood optimization function. Then, the estimated ego-motion maximizes this function. Large-scale real world experiments in an urban environment using a 77 GHz radar show the robust and accurate ego-motion estimation of the proposed algorithm. In the experiments, comparisons are made to state-of-the-art algorithms, the vehicle odometry, and a high-precision inertial measurement unit.

Clustering improved grid map registration using the normal distribution transform

Grid map registration is an important field in mobile robotics. Applications in which multiple robots are involved benefit from multiple aligned grid maps as they provide an efficient exploration of the environment in parallel. In this paper, a normal distribution transform (NDT)-based approach for grid map registration is presented. For simultaneous mapping and localization approaches on laser data, the NDT is widely used to align new laser scans to reference scans. The original grid quantization-based NDT results in good registration performances but has poor convergence properties due to discontinuities of the optimization function and absolute grid resolution. This paper shows that clustering techniques overcome disadvantages of the original NDT by significantly improving the convergence basin for aligning grid maps. A multi-scale clustering method results in an improved registration performance which is shown on real world experiments on radar data.

Landmark based radar SLAM using graph optimization

On the way to achieving higher degrees of autonomy for vehicles in complicated, ever changing scenarios, the localization problem poses a very important role. Especially the Simultaneous Localization and Mapping (SLAM) problem has been studied greatly in the past. For an autonomous system in the real world, we present a very cost-efficient, robust and very precise localization approach based on GraphSLAM and graph optimization using radar sensors. We are able to prove on a dynamically changing parking lot layout that both mapping and localization accuracy are very high. To evaluate the performance of the mapping algorithm, a highly accurate ground truth map generated from a total station was used. Localization results are compared to a high precision DGPS/INS system. Utilizing these methods, we can show the strong performance of our algorithm.

FSCD and BASD: Robust landmark detection and description on radar-based grids

This paper presents a detector and descriptor combination for robust landmarks on grids from radar measurements. Landmarks are fundamental for localization, a task of great importance in the field of mobile robotics and essential for autonomous driving. A fast detector is proposed which uses a rotational invariant pattern to locate scattering centers. These scattering centers occur as local maxima on a measurement grid, where detections from radar sensors for each cell are incremented. For association, a binary version of a descriptor designed especially for radar data is used. Experiments show that for radar data, the proposed combination improves performance compared to state-of-the-art algorithms.

Probabilistic rectangular-shape estimation for extended object tracking

This paper presents new methods for the representation of a vehicles contour by an oriented rectangle, also known as the bounding box. The parameters of this bounding box are originally modeled probabilistically by a single multivariate Gaussian distribution. This approach incorporates the sensor uncertainties, where the problem of estimating the parameters of this distribution from range measurements is addressed. Additionally, a transformation of the parameters into the measurement space is introduced. This representation is used to perform probabilistic updates by new measurements. The proposed method can handle strong parameter changes which might be affected by object occlusion. Experiments on real-world data demonstrate the robustness and accuracy of the probabilistic approach integrated in a tracking framework incorporating the Doppler measurements of automotive radars and laser measurements.

Efficient Dictionary Learning with Sparseness-Enforcing Projections

Learning dictionaries suitable for sparse coding instead of using engineered bases has proven effective in a variety of image processing tasks. This paper studies the optimization of dictionaries on image data where the representation is enforced to be explicitly sparse with respect to a smooth, normalized sparseness measure. This involves the computation of Euclidean projections onto level sets of the sparseness measure. While previous algorithms for this optimization problem had at least quasi-linear time complexity, here the first algorithm with linear time complexity and constant space complexity is proposed. The key for this is the mathematically rigorous derivation of a characterization of the projection’s result based on a soft-shrinkage function. This theory is applied in an original algorithm called Easy Dictionary Learning (EZDL), which learns dictionaries with a simple and fast-to-compute Hebbian-like learning rule. The new algorithm is efficient, expressive and particularly simple to implement. It is demonstrated that despite its simplicity, the proposed learning algorithm is able to generate a rich variety of dictionaries, in particular a topographic organization of atoms or separable atoms. Further, the dictionaries are as expressive as those of benchmark learning algorithms in terms of the reproduction quality on entire images, and result in an equivalent denoising performance. EZDL learns approximately 30 % faster than the already very efficient Online Dictionary Learning algorithm, and is therefore eligible for rapid data set analysis and problems with vast quantities of learning samples.

Probabilistic ego-motion estimation using multiple automotive radar sensors

Abstract For automotive applications, an accurate estimation of the ego-motion is required to make advanced driver assistant systems work reliably. The proposed framework for ego-motion estimation involves two components: The first component is the spatial registration of consecutive scans. In this paper, the reference scan is represented by a sparse Gaussian Mixture model. This structural representation is improved by incorporating clustering algorithms. For the spatial matching of consecutive scans, a normal distributions transform-based optimization is used. The second component is a likelihood model for the Doppler velocity. Using a hypothesis for the ego-motion state, the expected radial velocity can be calculated and compared to the actual measured Doppler velocity. The ego-motion estimation framework of this paper is a joint spatial and Doppler-based optimization function which shows reliable performance on real world data and compared to state-of-the-art algorithms.