RoadGraph: High level sensor data fusion between objects and street network

Homeier, Kai; Wolf, Lars

doi:10.1109/itsc.2011.6082803

Cited by 17 publications

(6 citation statements)

References 1 publication

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We consider 15 manually engineered features. Infrastructural features are derived from street networks [22]. The vehicle kinematics are described by derivatives of lateral and longitudinal actions.…”

Section: B Reward Feature Representationmentioning

confidence: 99%

Planning on the fast lane: Learning to interact using attention mechanisms in path integral inverse reinforcement learning

Rosbach¹,

Li²,

Großjohann³

et al. 2020

2020 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

View full text Add to dashboard Cite

General-purpose trajectory planning algorithms for automated driving utilize complex reward functions to perform a combined optimization of strategic, behavioral, and kinematic features. The specification and tuning of a single reward function is a tedious task and does not generalize over a large set of traffic situations. Deep learning approaches based on path integral inverse reinforcement learning have been successfully applied to predict local situation-dependent reward functions using features of a set of sampled driving policies. Sample-based trajectory planning algorithms are able to approximate a spatio-temporal subspace of feasible driving policies that can be used to encode the context of a situation. However, the interaction with dynamic objects requires an extended planning horizon, which requires sequential context modeling. In this work, we are concerned with the sequential reward prediction over an extended time horizon. We present a neural network architecture that uses a policy attention mechanism to generate a low-dimensional context vector by concentrating on trajectories with a human-like driving style. Besides, we propose a temporal attention mechanism to identify context switches and allow for stable adaptation of rewards. We evaluate our results on complex simulated driving situations, including other vehicles. Our evaluation shows that our policy attention mechanisms learns to focus on collision free policies in the configuration space. Furthermore, the temporal attention mechanism learns persistent interaction with other vehicles over an extended planning horizon.

show abstract

Section: B Reward Feature Representationmentioning

confidence: 99%

Planning on the fast lane: Learning to interact using attention mechanisms in path integral inverse reinforcement learning

Rosbach¹,

Li²,

Großjohann³

et al. 2020

2020 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

View full text Add to dashboard Cite

show abstract

“…3g. Infrastructural features are derived by a data fusion between objects and street network [21]. The kinematic characteristics of the policies are given by derivatives of the lateral and longitudinal actions as well as features related to behavior, e.g.…”

Section: B Reward Feature Representationmentioning

confidence: 99%

Driving Style Encoder: Situational Reward Adaptation for General-Purpose Planning in Automated Driving

Rosbach,

James,

Großjohann

et al. 2019

Preprint

View full text Add to dashboard Cite

General-purpose planning algorithms for automated driving combine mission, behavior, and local motion planning. Such planning algorithms map features of the environment and driving kinematics into complex reward functions. To achieve this, planning experts often rely on linear reward functions. The specification and tuning of these reward functions is a tedious process and requires significant experience. Moreover, a manually designed linear reward function does not generalize across different driving situations. In this work, we propose a deep learning approach based on inverse reinforcement learning that generates situation-dependent reward functions. Our neural network provides a mapping between features and actions of sampled driving policies of a modelpredictive control-based planner and predicts reward functions for upcoming planning cycles. In our evaluation, we compare the driving style of reward functions predicted by our deep network against clustered and linear reward functions. Our proposed deep learning approach outperforms clustered linear reward functions and is at par with linear reward functions with a-priori knowledge about the situation.

show abstract

“…An example of the operational level is given by object lists with hypotheses of traffic participants perceived by infrastructure (e.g. Homeier and Wolf [22]). …”

Section: External Data 431 Processed Datamentioning

confidence: 99%

Autonomous driving – a top-down-approach

Matthaei

Maurer

2015

At - Automatisierungstechnik

View full text Add to dashboard Cite

This paper presents a functional system architecture for an "autonomous vehicle" in the sense of a modular building block system. It is developed in a topdown approach based on the definition of the functional requirements for an autonomous vehicle and explicitly combines perception-based and localization-based approaches. Both the definition and the functional system architecture consider the aspects operating by the human being, mission accomplishment, map data, localization, environmental and self-perception as well as cooperation.The functional system architecture is developed in the context of the research project "Stadtpilot" at the Technische Universität Braunschweig. Zusammenfassung:In diesem Artikel stellen wir eine funktionale Systemarchitektur für ein "autonom fahrendes Straßenfahrzeug" vor, die im Sinne eines modularen Baukastensystems entworfen ist. Sie wurde in einem Top-Down-Ansatz ausgehend von einer Definition des Funktionsumfangs eines "autonom fahrenden Straßen-fahrzeugs" entwickelt und führt explizit wahrnehmungsbasierte und lokalisierungsbasierte Ansätze zusammen. Sowohl die Definition des Funktionsumfanges als auch die funktionale Systemarchitektur berücksichtigen die Aspekte Bedienung, Missionsumsetzung, Karten, Lokalisierung, Umfeld-und Selbstwahrnehmung sowie Kooperation. Die Ergebnisse basieren unter anderem auf Erkenntnissen aus dem Projekt "Stadtpilot" der Technischen Universität Braunschweig.

show abstract

RoadGraph: High level sensor data fusion between objects and street network

Cited by 17 publications

References 1 publication

Planning on the fast lane: Learning to interact using attention mechanisms in path integral inverse reinforcement learning

Planning on the fast lane: Learning to interact using attention mechanisms in path integral inverse reinforcement learning

Driving Style Encoder: Situational Reward Adaptation for General-Purpose Planning in Automated Driving

Autonomous driving – a top-down-approach

Contact Info

Product

Resources

About