Periodic Reporting for period 1 - COSAFE (Cooperative Connected Intelligent Vehicles for Safe and Efficient Road Transport)
Período documentado: 2018-11-01 hasta 2020-10-31
Road accidents and traffic congestion are serious problems for global transport systems. Connected vehicles (CV) and automated vehicles (AV) are among the most heavily researched automotive technologies to reduce road accidents and improve road efficiency. They have paramount safety, economic, road efficiency and mobility benefits. However, both AV and CV technologies have inherent shortcomings. In view of the limitations of CV and AV, the COSAFE project investigates the technology of connected intelligent vehicles (CIV). The idea is motivated by the observation that vehicles are increasingly intelligent with various levels of autonomous functionalities, which are enabled by more sensing and computing resources. The massive sensor and computing resources of vehicles and the transport infrastructure could be shared and exploited through effective cooperation of the CIVs to improve road safety and efficiency (RSE). Sensing information that can be shared include driving paths, moving objects within the drivable area, scene semantics (such as traffic lights, traffic signs and on-road markings) and driving manuverments (e.g. turning and giving way), etc. With resource sharing and cooperation, connected intelligent vehicles can have comprehensive perception of driving environments; innovative technologies, applications and systems could be developed to reduce road accidents and improve transport network efficiency. To our best knowledge, there is no research reported on the CIV technologies and applications.
While CIV holds great potentials, there are still many challenges and unanswered questions, for example unclear road safety impact of CIV with various levels of vehicle intelligence, security and safety issues for resource sharing and cooperation, and vehicles networking issue. This COSAFE RISE project directly addresses these issues by developing innovative vehicular networking and resource sharing and cooperation mechanisms, and integrating them into CIV to develop cooperative CIV technologies and applications.
The overall scientific R&I objectives (RO) of this COSAFE project include:
- RO-1: Identify the key impact, opportunities and challenges of CIV for road transport networks.
- RO-2: Research and develop reliable & high capacity vehicular communication and networking solutions with high reliability and capacity under dynamic mobile network scenarios.
- RO-3: Research and design scalable, effective and secure cooperative sensing and computing (CSC) mechanisms and algorithms for CIVs with heterogeneous sensing and computing resources.
- RO-4: Develop and pilot test novel cooperative CIV technologies and applications with robust CSC to reduce road accidents and improve road efficiency.
- RO-5: Understand the traffic pattern of CIV applications and quantify its impact on V2X communication performance, develop effective CIV application traffic prediction method, and develop service-oriented network resource management schemes to maximize resource utilization and RSE performance.
While CIV holds great potentials, there are still many challenges and unanswered questions, for example unclear road safety impact of CIV with various levels of vehicle intelligence, security and safety issues for resource sharing and cooperation, and vehicles networking issue. This COSAFE RISE project directly addresses these issues by developing innovative vehicular networking and resource sharing and cooperation mechanisms, and integrating them into CIV to develop cooperative CIV technologies and applications.
The overall scientific R&I objectives (RO) of this COSAFE project include:
- RO-1: Identify the key impact, opportunities and challenges of CIV for road transport networks.
- RO-2: Research and develop reliable & high capacity vehicular communication and networking solutions with high reliability and capacity under dynamic mobile network scenarios.
- RO-3: Research and design scalable, effective and secure cooperative sensing and computing (CSC) mechanisms and algorithms for CIVs with heterogeneous sensing and computing resources.
- RO-4: Develop and pilot test novel cooperative CIV technologies and applications with robust CSC to reduce road accidents and improve road efficiency.
- RO-5: Understand the traffic pattern of CIV applications and quantify its impact on V2X communication performance, develop effective CIV application traffic prediction method, and develop service-oriented network resource management schemes to maximize resource utilization and RSE performance.
Since the start of the project, the COSAFE partners have conducted research and innovation activites according to the proposed research directions and plan. More specifically, major efforts are dedicated to work package 2, which is dedicated to improve the reliability and capacity of V2X networks, in order to support effectively the high layer cooperative sensing and computing and the cooperative safety and efficiency applications.
The main research works included development and application of context for reliable V2X networking, high capacity V2X networks with millimetre wave and visible light communication, testbed development and trials. The partners involved in WP2 have successfully achieved the research targets set for the first stage of the research for WP2, publishing more than five high quality research papers.
The main results include:
- 5G millimetre channel model for V2X communication and broadcast algorithms are proposed and developed for high capacity and reliable vehicular communications.
- Wideband spectrum sensing is studied to provide higher capacity and more reliable communication with dynamic access to white space spectrum resources.
- Visible Light Communication (VLC) techniques are exploited and studied for intelligent transportation system solution to safely coordinate autonomous vehicles crossing at roundabouts.
- The development and application of network and application contexts for reliable and efficient V2X broadcast communication schemes are investigated.
- A cost effective, scalable and flexible software defined radio and Dedicated Short-Range Communications (DSRC) based V2V testbed is proposed and tested, which can be used for safe and efficient cooperative intelligent transport systems.
The main research works included development and application of context for reliable V2X networking, high capacity V2X networks with millimetre wave and visible light communication, testbed development and trials. The partners involved in WP2 have successfully achieved the research targets set for the first stage of the research for WP2, publishing more than five high quality research papers.
The main results include:
- 5G millimetre channel model for V2X communication and broadcast algorithms are proposed and developed for high capacity and reliable vehicular communications.
- Wideband spectrum sensing is studied to provide higher capacity and more reliable communication with dynamic access to white space spectrum resources.
- Visible Light Communication (VLC) techniques are exploited and studied for intelligent transportation system solution to safely coordinate autonomous vehicles crossing at roundabouts.
- The development and application of network and application contexts for reliable and efficient V2X broadcast communication schemes are investigated.
- A cost effective, scalable and flexible software defined radio and Dedicated Short-Range Communications (DSRC) based V2V testbed is proposed and tested, which can be used for safe and efficient cooperative intelligent transport systems.
The COSAFE project moves beyond the state of the art by developing innovative cooperative CIV technologies and applications with key supporting technologies to provide efficient RSE solutions. By bringing together academics, industry and local policymaker, through effective secondments and knowledge sharing activities, COSAFE could a achieve the following expected results:
- Novel context aware V2X communication and networking algorithms that can deliver high reliability and low latency for CIV applications under dynamic mobile network scenarios. Targeting packet reception ratio (PRR) of 0.9 for collaborative safety applications, the proposed solution should increase effective communication range and support reliable exchange of additional sensing information under representative scenarios.
- A standardised dataset for cooperative sensing and data fusion with clear interface for public access, which will support benchmarking and advance research on cooperative sensing and cooperative CIV applications. Existing public benchmarking datasets such as KITTI datasets do not support cooperative sensing and CIV.
- Developing scalable, effective and secure CSC mechanisms and algorithms for CIV. Through innovative actions including robust signalling mechanisms, blockchain based trust and motivation model, and deep learning object detection models, the sensing range and NLoS problems of existing sensing technologies could be solved. The effective sensing range of ADAS systems could be extended.
- Novel traffic model and prediction for CIV applications and service-oriented network resource management schemes, which will be built on transport and communication network traffic monitoring and advanced cellular network resource allocation. This could significantly improve network resource utilization and CIV applications performance.
With the innovative cooperative CIV technologies and tools with robust CSC, more reaction time will be provided, road accidents will be reduced and road efficiency will be improved. The CSC based solutions will also impact wider scientific community, find wide application to intelligent transport systems, smart cities, Internet of Things and environment sectors.
- Novel context aware V2X communication and networking algorithms that can deliver high reliability and low latency for CIV applications under dynamic mobile network scenarios. Targeting packet reception ratio (PRR) of 0.9 for collaborative safety applications, the proposed solution should increase effective communication range and support reliable exchange of additional sensing information under representative scenarios.
- A standardised dataset for cooperative sensing and data fusion with clear interface for public access, which will support benchmarking and advance research on cooperative sensing and cooperative CIV applications. Existing public benchmarking datasets such as KITTI datasets do not support cooperative sensing and CIV.
- Developing scalable, effective and secure CSC mechanisms and algorithms for CIV. Through innovative actions including robust signalling mechanisms, blockchain based trust and motivation model, and deep learning object detection models, the sensing range and NLoS problems of existing sensing technologies could be solved. The effective sensing range of ADAS systems could be extended.
- Novel traffic model and prediction for CIV applications and service-oriented network resource management schemes, which will be built on transport and communication network traffic monitoring and advanced cellular network resource allocation. This could significantly improve network resource utilization and CIV applications performance.
With the innovative cooperative CIV technologies and tools with robust CSC, more reaction time will be provided, road accidents will be reduced and road efficiency will be improved. The CSC based solutions will also impact wider scientific community, find wide application to intelligent transport systems, smart cities, Internet of Things and environment sectors.