Automated driving of Porsche Panamera

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21730%2F19%3A00339797" target="_blank" >RIV/68407700:21730/19:00339797 - isvavai.cz</a>

Výsledek na webu

<a href="http://industrialinformatics.fel.cvut.cz/" target="_blank" >http://industrialinformatics.fel.cvut.cz/</a>

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

Automated driving of Porsche Panamera

Popis výsledku v původním jazyce

This work presents the algorithms and system architecture of an automated car aimed at performing a slalom use-case. Besides the sensors integrated into the Panamera car, the system is equipped with additional Differential GPS, mono camera, and 16-layers Lidar. The algorithms run on NVIDIA TX2, they are integrated using ROS middleware and interfaced with the car ECUs via FlexRay. HOG descriptor is used to create a feature vector, and SVM is used for cones classification. Data from the camera, Lidar, and DGPS are used to localize the cones and generate the map. Kalman filter derives an accurate car position and heading from DGPS and the car odometry. The bicycle model is used to formulate a nonlinear quadratic optimization problem aiming at optimal trajectory planning respecting the car kinematics. Finally, the trajectory following and lateral & longitudinal control are driving the car in the slalom. Substantial work was devoted to the experiments with a real vehicle and the fine-tuning of the system parameters. Validation of the system reveals exciting observations related to the precision, frequency, and sensitivity of the system components.

Název v anglickém jazyce

Automated driving of Porsche Panamera

Popis výsledku anglicky

This work presents the algorithms and system architecture of an automated car aimed at performing a slalom use-case. Besides the sensors integrated into the Panamera car, the system is equipped with additional Differential GPS, mono camera, and 16-layers Lidar. The algorithms run on NVIDIA TX2, they are integrated using ROS middleware and interfaced with the car ECUs via FlexRay. HOG descriptor is used to create a feature vector, and SVM is used for cones classification. Data from the camera, Lidar, and DGPS are used to localize the cones and generate the map. Kalman filter derives an accurate car position and heading from DGPS and the car odometry. The bicycle model is used to formulate a nonlinear quadratic optimization problem aiming at optimal trajectory planning respecting the car kinematics. Finally, the trajectory following and lateral & longitudinal control are driving the car in the slalom. Substantial work was devoted to the experiments with a real vehicle and the fine-tuning of the system parameters. Validation of the system reveals exciting observations related to the precision, frequency, and sensitivity of the system components.

Druh

R - Software

CEP obor

—

OECD FORD obor

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Projekt

<a href="/cs/project/TN01000024" target="_blank" >TN01000024: Národní centrum kompetence - Kybernetika a umělá inteligence</a><br>

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Rok uplatnění

2019

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Interní identifikační kód produktu

Panamera_slalom

Technické parametry

camera, lidar, ROS, trajectory planning, controllers, Kalman filter

Ekonomické parametry

Výsledek přispěl ke zvýšení zisku vlastníka výsledku a umožní navázání dalších kontraktů v oblasti automatizovaných automobilů.

IČO vlastníka výsledku

68407700

Název vlastníka

ČVUT CIIRC, CIIRC