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

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

doi:10.1109/iv51971.2022.9827360

Cited by 14 publications

(12 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It helps in extracting the information from LiDAR data and image processing that is required to populate grid cells. A multitask recurrent neural network is proposed to predict grid maps [30]. Grid maps provide sematic information, occupancies, velocity estimates, and drivable area.…”

Section: Perceptionmentioning

confidence: 99%

“…The maximum advantage of these learning controllers can be achieved as they make use of a modelbased control as well as learning algorithms. Deep learning-based techniques have gained much importance in the motion control of autonomous vehicles [30,65]. A visual attention model is used to train an end-to-end (from images to control commands) convolutional neural network model [66].…”

Section: Motion Controllers/actmentioning

confidence: 99%

See 1 more Smart Citation

Intelligence of Autonomous Vehicles: A Concise Revisit

et al. 2022

View full text Add to dashboard Cite

Artificial intelligence- (AI-) empowered machines are devised to mimic human actions. In the automotive industry, AI plays a significant role in the development of vehicular technology. AI joins hands with the field of mechatronics to assist in the accurate execution of the vehicle functionalities. Autonomous vehicles get the scene information by using onboard sensors such as laser, radar, lidar, Global Positioning System (GPS), and vehicular communication networks. The data obtained is then used for various path planning and control techniques to make the vehicles capable of autonomously driving in complex environments. Autonomous vehicles use very up-to-date AI algorithms to localize themselves in known and unknown environments. AI algorithms are also exploited for perception, path planning, and motion control. A concise review of the state-of-the-art techniques to improve the performance of autonomous vehicles is presented.

show abstract

Section: Perceptionmentioning

confidence: 99%

Section: Motion Controllers/actmentioning

confidence: 99%

Intelligence of Autonomous Vehicles: A Concise Revisit

et al. 2022

View full text Add to dashboard Cite

show abstract

“…Schreiber et al [19] apply recurrent neural networks to predict a dynamic occupancy grid map from measurement grids with an ego-motion compensation. An extension [3] uses raw lidar point clouds as input and predicts additional semantic labels for the cells. While grid maps can provide information about free space for driving and other traffic participant attributes, like orientation, velocity, and type, they never included pedestrian body pose data.…”

Section: A Grid Mapsmentioning

confidence: 99%

“…To safely navigate, the vehicle has to fully recognize the surrounding environment. This can be done with different environment models like grid maps [3] and target lists [4]. Existing environment models are designed to display the location of other traffic participants like cars or pedestrians on the map and also show the driveable area but neglect the differing characteristics of pedestrians.…”

mentioning

confidence: 99%

“…Common approaches to model the vehicle's environment are grid maps or track lists from multi-object tracking (MOT) [9]. Grid maps can be used to describe not only static environments [10] but also surroundings with dynamic objects [11] and can even include the object's type [3]. In MOT, the objects detected in the current frame are associated with objects from earlier frames to get their position and motion data.…”

mentioning

confidence: 99%

See 1 more Smart Citation

A Spatio-Temporal Multilayer Perceptron for Gesture Recognition

Holzbock

Tsaregorodtsev

Dawoud

et al. 2022

2022 IEEE Intelligent Vehicles Symposium (IV)

View full text Add to dashboard Cite

Gesture recognition is essential for the interaction of autonomous vehicles with humans. While the current approaches focus on combining several modalities like image features, keypoints and bone vectors, we present neural network architecture that delivers state-of-the-art results only with body skeleton input data. We propose the spatio-temporal multilayer perceptron for gesture recognition in the context of autonomous vehicles. Given 3D body poses over time, we define temporal and spatial mixing operations to extract features in both domains. Additionally, the importance of each time step is re-weighted with Squeeze-and-Excitation layers. An extensive evaluation of the TCG and Drive&Act datasets is provided to showcase the promising performance of our approach. Furthermore, we deploy our model to our autonomous vehicle to show its realtime capability and stable execution.

show abstract

Radars for Autonomous Driving: A Review of Deep Learning Methods and Challenges

Srivastav,

Mandal

2023

IEEE Access

View full text Add to dashboard Cite

Radar is a key component of the suite of perception sensors used for safe and reliable navigation of autonomous vehicles. Its unique capabilities include high-resolution velocity imaging, detection of agents in occlusion and over long ranges, and robust performance in adverse weather conditions. However, the usage of radar data presents some challenges: it is characterized by low resolution, sparsity, clutter, high uncertainty, and lack of good datasets. These challenges have limited radar deep learning research. As a result, current radar models are often influenced by lidar and vision models, which are focused on optical features that are relatively weak in radar data, thus resulting in under-utilization of radar's capabilities and diminishing its contribution to autonomous perception. This review seeks to encourage further deep learning research on autonomous radar data by 1) identifying key research themes, and 2) offering a comprehensive overview of current opportunities and challenges in the field. Topics covered include early and late fusion, occupancy flow estimation, uncertainty modeling, and multipath detection. The paper also discusses radar fundamentals and data representation, presents a curated list of recent radar datasets, and reviews state-of-the-art lidar and vision models relevant for radar research.

show abstract

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

Cited by 14 publications

References 28 publications

Intelligence of Autonomous Vehicles: A Concise Revisit

Intelligence of Autonomous Vehicles: A Concise Revisit

A Spatio-Temporal Multilayer Perceptron for Gesture Recognition

Radars for Autonomous Driving: A Review of Deep Learning Methods and Challenges

Contact Info

Product

Resources

About