System Architecture of a Driverless Electric Car in the Grand Cooperative Driving Challenge

Abstract

This paper presents the complete system architecture of a connected driverless electric car designed to participate in the Grand Cooperative Driving Challenge 2016. One of the main goals of this challenge was to demonstrate the feasibility of multiple autonomous vehicles cooperating via wireless communications on public roads. Several complex cooperative scenarios were considered, including the merging of two lanes and cooperation at an intersection. We describe in some detail an implementation using the open-source PACPUS framework that successfully completed the different tasks in the challenge. Our description covers localization, mapping, perception, control, communication and the human-machine interface. Some experimental results recorded in real-time during the challenge are reported.

Benchmarking Particle Filter Algorithms for Efficient Velodyne-Based Vehicle Localization

“…The integration of the localization techniques into vehicle systems’ architecture demands more computationally efficient techniques [14,15]. When high level tasks are demanded, as cooperative driving, real-time performance is critical [16]. Rao-Blackwellized particle filters (RBPF) have been proposed for Velodyne-based SLAM in [17], using a pre-processing stage where vertical objects are detected in the raw scans such that RBPF-SLAM can be fed with a reduced number of discrete features.…”

A Framework for the Synergistic Integration of Fully Autonomous Ground Vehicles With Smart City

“…HD maps with lane-level accuracy are a topic of particular interest to both map providers and car manufacturers [87]. Detailed static 3D HD map for high-precision position fixes down to a margin of error to 10 cm can be created today using real-time kinematic capabilities [27] and this is ten times as accurate as 2D maps that operate with a margin of error of up to a metre [9].…”

“…FAGVs using this static modelling involving the map of static obstacles in memory do not need to see the lanes and static obstacles on the roads. Furthermore, with Real-Time Kinematic (RTK) corrections, the solution combining a Global Navigation Satellite System (GNSS) and an Inertial Navigation System (INS) solution can provide both absolute accuracy and continuity for vehicle localisation [87]. More explicitly, the solution provides information on localisation (latitude, longitude, altitude), velocity (w.r.t.…”

“…More explicitly, the solution provides information on localisation (latitude, longitude, altitude), velocity (w.r.t. east, north, up directions), acceleration (lateral, longitudinal, vertical), rotation (roll, pitch, azimuth) and rotation rate (roll rate, pitch rate, yaw rate) by providing the standard deviations for all these estimates [87]. Using these static and dynamic localisation approaches based on Geodetic coordinates (i.e., latitude and longitude) and Cartesian coordinates, FAGVs can travel from one location to another without using other external sensors if there were not any mobile objects around those vehicles.…”

A Novel Design and Implementation of Autonomous Robotic Car Based on ROS in Indoor Scenario

“…For example, some outstanding companies, such as Tesla, Google and Baidu, have successfully deployed autonomous cars in cities, and these cars are expected to be deployed in suburbs of city or rural areas. On top of achievements from leading corporations, numerous autonomous vehicle schemes from universities or individuals, either motor-based schemes [3][4][5][6][7][8][9][10][11] or electronic-based schemes [12][13][14][15][16][17][18][19], are also extraordinary and have decent performance in various of situations.…”