Dynamic Occupancy Grid Mapping with Recurrent Neural Networks

Schreiber, Marcel; Belagiannis, Vasileios; Gläser, Claudius; Dietmayer, Klaus

doi:10.1109/icra48506.2021.9561375

Cited by 39 publications

(24 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The estimation is applied to generate non-stationary kernels in the Hilbert space to build the dynamic Hilbert map. With the popularity of deep learning methods, some recent works [24], [25] adopt recurrent neural networks to predict the velocity of each grid in a grid map.…”

Section: B Occupancy Maps In Dynamic Environmentsmentioning

confidence: 99%

“…Compared to the above mentioned methods [21], [23], [22], [24], [25], particle-based methods are originally designed for dynamic environments. In particle-based methods, an obstacle is regarded as a set of point objects.…”

Section: Particle-based Dynamic Occupancy Mapsmentioning

confidence: 99%

“…The complexity of the update step in (33) and ( 34) is then about (2n+1) 2 N f times to Equations ( 24) and (25), where N f = θ h θv 2 , described in Section III, is the number of pyramid subspaces in S f . Suppose r min = 0.15m, σ = 1%, = 0.01 and the FOV is θ h = 90 • and θ v = 60 • .…”

Section: Let Z Smentioning

confidence: 99%

See 2 more Smart Citations

Continuous Occupancy Mapping in Dynamic Environments Using Particles

Chen¹,

Wang²,

Peng³

et al. 2022

Preprint

View full text Add to dashboard Cite

Dynamic occupancy maps were proposed in recent years to model the obstacles in dynamic environments. Among these maps, the particle-based map offers a solid theoretical basis and the ability to model complex-shaped obstacles. Current particle-based maps describe the occupancy status in discrete grid form and suffer from the grid size problem, namely: large grid size is unfavorable for path planning while small grid size lowers efficiency and causes gaps and inconsistencies. To tackle this problem, this paper generalizes the particle-based map into continuous space and builds an efficient 3D local map. A dual-structure subspace division paradigm, composed of a voxel subspace division and a novel pyramid-like subspace division, is proposed to propagate particles and update the map efficiently with the consideration of occlusions. The occupancy status of an arbitrary point can then be estimated with the cardinality expectation. To reduce the noise in modeling static and dynamic obstacles simultaneously, an initial velocity estimation approach and a mixture model are utilized. Experimental results show that our map can effectively and efficiently model both dynamic obstacles and static obstacles. Compared to the state-of-the-art grid-form particle-based map, our map enables continuous occupancy estimation and substantially improves the performance in different resolutions. We also deployed the map on a quadrotor to demonstrate the bright prospect of using this map in obstacle avoidance tasks of small-scale robotics systems.

show abstract

Section: B Occupancy Maps In Dynamic Environmentsmentioning

confidence: 99%

Section: Particle-based Dynamic Occupancy Mapsmentioning

confidence: 99%

See 1 more Smart Citation

Continuous Occupancy Mapping in Dynamic Environments Using Particles

Chen¹,

Wang²,

Peng³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…The classical method for static environments is occupancy grid mapping [1][2][3] where maps are divided into a grid and the states of different grid cells are assumed to be independent. In dynamic environments, one popular strategy is to estimate the number of potential targets, their positions, and velocities from sensor data [4][5][6]. The dynamic object detection needs to identify the objects and their correspondence in different time instants.…”

Section: Introduction 1literature Reviewmentioning

confidence: 99%

HMM-Based Dynamic Mapping with Gaussian Random Fields

Barão

Rato

et al. 2022

Electronics

View full text Add to dashboard Cite

This paper focuses on the mapping problem for mobile robots in dynamic environments where the state of every point in space may change, over time, between free or occupied. The dynamical behaviour of a single point is modelled by a Markov chain, which has to be learned from the data collected by the robot. Spatial correlation is based on Gaussian random fields (GRFs), which correlate the Markov chain parameters according to their physical distance. Using this strategy, one point can be learned from its surroundings, and unobserved space can also be learned from nearby observed space. The map is a field of Markov matrices that describe not only the occupancy probabilities (the stationary distribution) as well as the dynamics in every point. The estimation of transition probabilities of the whole space is factorised into two steps: The parameter estimation for training points and the parameter prediction for test points. The parameter estimation in the first step is solved by the expectation maximisation (EM) algorithm. Based on the estimated parameters of training points, the parameters of test points are obtained by the predictive equation in Gaussian processes with noise-free observations. Finally, this method is validated in experimental environments.

show abstract

“…In this work, we introduce a multi-task recurrent neural network to predict dynamic occupancy grid maps with semantic classes for the occupied cells, based on raw lidar data to tackle these drawbacks. The multi-task network architecture is based on the approach in [5], but extended with an additional decoder for the semantic classification and omitting the calculation of measurement grid maps as input data. The input data in this work is a discretization of the 3D lidar data in a bird's eye view (BEV) with several channels to represent the height.…”

Section: Introductionmentioning

confidence: 99%

A Multi-Task Recurrent Neural Network for End-to-End Dynamic Occupancy Grid Mapping

Schreiber¹,

Belagiannis²,

Gläser³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

A common approach for modeling the environment of an autonomous vehicle are dynamic occupancy grid maps, in which the surrounding is divided into cells, each containing the occupancy and velocity state of its location. Despite the advantage of modeling arbitrary shaped objects, the used algorithms rely on hand-designed inverse sensor models and semantic information is missing. Therefore, we introduce a multi-task recurrent neural network to predict grid maps providing occupancies, velocity estimates, semantic information and the driveable area. During training, our network architecture, which is a combination of convolutional and recurrent layers, processes sequences of raw lidar data, that is represented as bird's eye view images with several height channels. The multi-task network is trained in an end-to-end fashion to predict occupancy grid maps without the usual preprocessing steps consisting of removing ground points and applying an inverse sensor model. In our evaluations, we show that our learned inverse sensor model is able to overcome some limitations of a geometric inverse sensor model in terms of representing object shapes and modeling freespace. Moreover, we report a better runtime performance and more accurate semantic predictions for our end-to-end approach, compared to our network relying on measurement grid maps as input data.

show abstract

Dynamic Occupancy Grid Mapping with Recurrent Neural Networks

Cited by 39 publications

References 38 publications

Continuous Occupancy Mapping in Dynamic Environments Using Particles

Continuous Occupancy Mapping in Dynamic Environments Using Particles

HMM-Based Dynamic Mapping with Gaussian Random Fields

A Multi-Task Recurrent Neural Network for End-to-End Dynamic Occupancy Grid Mapping

Contact Info

Product

Resources

About