An ETSI ITS-enabled Robotic Scale Testbed for Network-Aided Safety-Critical Scenarios

• Type: Article in International Conference Proceedings Book
• Date: 2023

Authors

Pinheiro, JM (Author) Other; Filho, EV (Author) Other; Santos, PM (Author) Other; Luis Almeida (Author) FEUP

Conference proceedings International

Title: 2023 53RD ANNUAL IEEE/IFIP INTERNATIONAL CONFERENCE ON DEPENDABLE SYSTEMS AND NETWORKS WORKSHOPS, DSN-W

Pages: 52-59

53rd Annual IEEE/IFIP International Conference on Dependable Systems and Networks (DSN)

Porto, PORTUGAL, JUN 27-30, 2023

Indexing

ISI Web of Knowledge - 0 Citations

Scopus - 1 Citation

Other information

Authenticus ID: P-00Y-ZRM

DOI: 10.1109/dsn-w58399.2023.00030

Abstract (EN): Vehicle-to-Everything (V2X) communications allows new and exciting relevant Intelligent Transportation Systems (ITS) applications such as cooperative perception. In some vehicular use-cases, infrastructure can have a crucial role in safeguarding safety, e.g. in an intersection in which inflowing vehicles do not have Line-of-Sight (blind corner) and an emergency braking action is required. Performance evaluation of vehicular communications systems (end-to-end delay, packet loss ratio, etc.) often resorts to static wireless testbeds, but there tends to be little consideration about the other subsystems involved in the time-sensitive goal, namely sensors, processing and decision-making, and vehicle actuators. In this paper we present a laboratorial testbed in which ETSI ITS/IEEE 802.11p-based On-Board and Road-Side Units (OBU/RSU) are deployed on a 1/10-scale autonomous robotic vehicle and on the road-side respectively, the latter being part of a road-side infrastructure that includes a camera and an edge processing node. We target a use-case of collision avoidance supported by the network, in which the infrastructure detects an impending collision and issues a DEN message to prevent it. We aim to take a step beyond the traditional end-to-end delay characterization that is limited to the communication subsystem, by proposing a tool that enables realistic characterization of the entire end-to-end (detection-to-action) delay in safety-critical use-cases. Results show that the end-to-end latency of the system (camera-edge processing node-RSU-OBU-vehicle actuators) is under 100ms.

Language: English

Type (Professor's evaluation): Scientific

No. of pages: 8