- Wireless Comms
- Bat System
- Broadband Phone
- Sentient Vehicles
- Task Assignment
- Computing for the Future of the Planet
- Active Floor
- Open-Source CSK Energy
- Research Overview
- MPhil Information
DTG Local Pages
- Cambridge Weather
- Contact Details
Sentient Vehicles & Sentient Transportation
Indoor sensor systems are typically static, and tailored to their environment. But what happens when the sensor platform becomes mobile? The Sentient vehicle subproject considers the integration of motoring and sensor systems, asking how can a reliable communication link be established? What data should be sent? What benefits can be identified? What risks are associated? Intelligent transportation systems are growing at a significant rate, and it is important to identify early-on the possible applications of such systems and their requirements. This aim is also present in the Transport Information Monitoring Environment (TIME) project.
Further information about the arrangements for the van's use can be found on the Sentient Van page.
We maintain a list of car-related research projects elsewhere. If you have links that would be useful to add, please let David (dnc25) or Jon (jjd27) know.
A write-up on the van and our ideas on privacy sensitive congestion charging, as presented at the Cambridge-MIT Institute's Connected Car seminar may also be of interest.
In order to research these questions, the DTG has fitted out a van with a variety of sensors and computing equipment. The sensors include:
- 2 GPS receivers
- OBD-II interface (for data from the vehicle's CAN bus)
- RFID identification card reader
- Digital video cameras
- Carbon dioxide sensor
- Temperature sensor
- Humidity sensor
- Barometric pressure sensor
- Inclination sensor
- 2-axis Accelerometer
- 2-axis Magnetometer (for sensing heading)
We also have various communication interfaces:
- IEEE 802.11b/g wireless interface
- IEEE 802.11a wireless interface (the technology that the IEEE Wireless Access in Vehicular Environments standard is based on)
- GSM GPRS & UMTS 3G cellular interface
- We are in the midst of a project to deploy an IEEE 802.16 WiMAX network
Our system is designed to be fully extensible, providing a scalable platform on which sentient applications can be deployed, and research into intelligent transportation carried out. We have already been able to collaborate with another group in the Computer Laboratory by providing them with videos of drivers' faces.
- Congestion-Aware Traffic Routing. Vehicles upload data about the congestion they have experienced to public access points which can be queried by other vehicles.
- Driver Expression Inference. We have recorded videos of the face of the driver, to be used for research into real-time inference of the driver's mood.
- Updating Digital Maps. Up-to-date digital road maps can be automatically generated from GPS traces collected from vehicles.
- Performance of IEEE 802.11a. We are examining the performance of the technology selected for the WAVE standard for vehicle-to-roadside communication in urban environments.
- Wireless Coverage Mapping. We are using the vehicle's GPRS/UMTS data card to map the wireless coverage available in an urban environment.
- Atmospheric Chemistry. We use the vehicle's Carbon Dioxide sensor to measure atmospheric pollution. This will be used to confirm models produced from government-owned fixed sensor deployments.
- Congestion Charging. We have devised a novel scheme for congestion charging which is peer-enforced to provide better privacy for honest drivers than traditional schemes. We intend to use the vehicle to further investigate technologies for congestion charging.
- Vehicle-to-Vehicle Communications. We have equipped a second vehicle in a similar manner for use in investigating communication between vehicles at speed in an urban environment.
- WiMAX Deployment. We are deploying an IEEE 802.16 network over a large urban area for further evaluation of wireless technologies.
Publications arising from our research into sentient transportation are listed below, most recent first.
Language-Based Optimisation of Sensor-Driven Distributed Computing ApplicationsJonathan J. Davies, Alastair R. Beresford, Alan Mycroft
11th International Conference on Fundamental Approaches to Software Engineering (FASE 2008), LNCS 4961, pp 407--422. Springer-Verlag, March 2008.
In many distributed computing paradigms, especially sensor networks and ubiquitous computing but also grid computing and web services, programmers commonly tie their application to a particular set of processors. This can lead to poor utilisation of resources causing increased compute time, wasted network bandwidth or poor battery life, particularly if later changes to the architecture or application render early decisions inappropriate. This paper describes a system which separates application code from the description of the resources available to execute it. Our framework and prototype compiler determines the best location to execute different parts of the distributed application. In addition, our language encourages the programmer to structure data, and the operations performed on it, as monoids and monoid homomorphisms. This approach enables the compiler to apply particular program transformations in a semantically-safe way, and therefore further increase the flexibility of the assignment of application tasks to available resources.
Scalable Inter-Vehicular ApplicationsJonathan J. Davies, Alastair R. Beresford
On the Move to Meaningful Internet Systems 2007: OTM 2007 Workshops (Part II). LNCS 4806, pp 876--885. Springer-Verlag, November 2007.
Many pervasive inter-vehicular applications involve the collation, processing and summarisation of sensor data originating from vehicles. When and where such processing takes place is an explicit design-stage decision. Often some processing occurs on vehicles, and some on back-end servers, but it is hard for the programmer to optimise this distribution for feasibility or performance. This paper investigates automated task assignment: we define a computational model which captures data aggregation and summarisation explicitly, allowing a compiler to automatically optimise the assignment of processing tasks to particular vehicles and servers. Our model allows a compiler to apply program transformations to data processing, which can further improve task assignment.
Survey of Technologies for the Implementation of National-scale Road User ChargingDavid N. Cottingham, Alastair R. Beresford, Robert K. Harle
Transport Reviews, Vol. 27, No. 4, Routledge, July 2007 (pages 499--523)
This paper surveys the technologies available for constructing a pervasive, national-scale road pricing system. It defines the different types of road pricing, the methods by which a vehicle's position can be determined, and then examines possible pricing regimes in the context of their technological requirements and implications. The issue of enforcement and the distribution of pricing policies are considered, and further complexities are outlined. An examination of the security aspects of such systems is made, focusing particularly on the need to ensure privacy using technological, rather than solely procedural, methods. The survey concludes that a pervasive, national-scale deployment is unlikely to be technically achievable in the short term.
Author Posting. © Taylor & Francis, 2007. This is the author's version of the work. It is posted here by permission of Taylor & Francis for personal use, not for redistribution. The definitive version was published in Transport Reviews, Volume 27 Issue 4, July 2007. doi:10.1080/01441640701214304 ( http://dx.doi.org/10.1080/01441640701214304).
Performance of IEEE 802.11a in Vehicular ContextsDavid N. Cottingham, Ian J. Wassell, Robert K. Harle
Proceedings of the 65th IEEE Vehicular Technology Conference (VTC Spring 2007), April 2007, Dublin, Ireland (pages 854--858)
A key component of intelligent transportation is the provision of adequate network infrastructure to support vehicle-to-vehicle and vehicle-to-roadside communication. In this paper we report on performance evaluations carried out using the IEEE 802.11a protocol at 5.2 GHz between a moving vehicle and a fixed base station. We concentrate our evaluation on realistic urban speeds and environments, observing that performance at very low speeds is degraded due to the presence of null zones. We vary the modulation scheme and analyse the spread of resulting throughputs. Our results have implications for multimedia and other real-time applications that will utilise vehicle-to-roadside connectivity.
- PDF (175 KB) (Copyright 2007 IEEE.)
A Vision for Wireless Access on the Road NetworkDavid N. Cottingham, Jonathan J. Davies
Proceedings of the 4th International Workshop on Intelligent Transportation (WIT 2007), March 2007, Hamburg, Germany (pages 25--30)
Metropolitan area wireless networks are currently being deployed in major cities around the world, whilst in tandem there has been much research into vehicle-to-roadside communication. New applications for vehicular networking become possible as blanket, low-cost, wireless networks begin to exist across cities, resulting in Connected Traffic, rather than isolated Connected Cars. We classify these applications according to their distinguishing characteristics, and discuss their network architecture requirements. We outline our current work on the language and compiler support required in order to deploy applications on such networks, and how coverage mapping algorithms will enable better prediction of network conditions to optimise such deployments. We conclude that such analysis and tools are important for defining the future of wireless access for the road network.
A Sensor Platform for Sentient Transportation ResearchJonathan J. Davies, David N. Cottingham, Brian D. Jones
1st European Conference on Smart Sensing and Context (EuroSSC) 2006, Enschede, The Netherlands
(LNCS volume 4272, pages 226--229, 2006)
This paper describes experiences and lessons learnt in the creation of a vehicle-based sensor platform, as part of research into sentient computing. We outline the requirements of such a platform; the sensors that have been deployed in it; the on-board infrastructure to facilitate logging and the deployment of context-aware applications; and the external communications provision. We focus particularly on the lessons learnt in the deployment, justifying and evaluating our approach. This analysis will be of use to others considering similar sensor platform deployments. We conclude with an outline of our ongoing research.
Scalable, Distributed, Real-Time Map GenerationJonathan J. Davies, Alastair R. Beresford, Andy Hopper
IEEE Pervasive Computing Magazine, Volume 5, Number 4, Pages 47-54, Oct-Dec 2006
With the advent of vehicles with sufficient computing power and communications capabilities, a range of new applications involving many vehicles are becoming feasible. These applications will involve the large-scale collection, processing, and dissemination of data, in which vehicles could participate as the sources, processors, and sinks. One such application is the automatic generation of up-to-date digital road maps based on the collection and processing of vehicles' location data. However, deploying such an application will involve several challenges and issues, including the choice of architecture to support it.This article is part of a special issue on Intelligent Transportation.
A Research Platform for Sentient TransportDavid N. Cottingham, Jonathan J. Davies, Brian D. Jones
Works in Progress item in IEEE Pervasive Computing Magazine, Volume 5, Number 4, Pages 63--64, Oct-Dec 2006
Describes the sentient vehicles project at the Computer Laboratory.
Privacy-Sensitive Congestion ChargingAlastair R. Beresford, Jonathan J. Davies, Robert K. Harle
14th International Workshop on Security Protocols, Series LNCS, Springer-Verlag, March 2006
National-scale congestion charging schemes are increasingly viewed as the most viable long-term strategy for controlling congestion and maintaining the viability of the road network. In this paper we challenge the widely held belief that enforceable and economically viable congestion charging schemes require drivers to give up their location privacy to the government. Instead we explore an alternative scheme where privately-owned cars enforce congestion charge payments by using an on-board vehicle unit containing a camera and wireless communications. Our solution prevents centralised tracking of vehicle movements but raises an important issue: should we trust our neighbours with a little personal information in preference to entrusting it all to the government?
- [To appear]
The Sentient VanDavid N. Cottingham, Jonathan J. Davies, Brian D. Jones
Internal DTG Poster, December 2005
The sentient van project has created a vehicular platform into which many different sensors and applications are integrated, allowing re- search into Intelligent Transportation Systems (ITS), network con- nectivity on the move, and congestion charging. This poster overviews the infratructure deployed in the van, and the current research work being carried out.
Keeping Big Brother Off the RoadRobert K. Harle, Alastair R. Beresford
IEE Review, Volume 51, Issue 10, Pages 34--37, October 2005
Road pricing schemes dependent on massive centralised databases will be hugely expensive while posing a threat to personal liberty. This article proposes an alternative. In operation, vehicles would use a radio interface to communicate with the central charging authority, purchasing congestion charge tickets and downloading current prices and charging zones into a local database. Communication with other vehicles is also by radio. As the vehicle moves, the positioning system estimates the instantaneous location, which, used in combination with the local database, allows the computer to determine the current charging zone. The computer also analyses the video stream from a camera and uses automatic number plate recognition to identify any vehicles in front of it. If it successfully identifies a vehicle, the computer uses the radio channel to request the electronic ticket from the identified vehicle; thereby verifying payment. If verification fails, the image is stored by the car, until communication with the central authority becomes possible. In this scheme, enforcement is achieved primarily by other vehicles, and not the central authority.
Congestion-Aware Vehicular Traffic Routing Using WiFi HotspotsDavid N. Cottingham, Jonathan J. Davies, Alastair R. Beresford
Poster at the Communications Innovation Institute workshop, April 2005, Cambridge, UK
Congestion on roads in the UK costs the economy an estimated £12 billion a year, a figure that is set to rise with increasing commuting distances and distance selling volumes. In cities, congestion on key routes must be managed to ensure smooth traffic flow, even at peak times. We have investigated how vehicles' routes can be improved by using congestion data collected on the vehicles and distributed by wireless access points. Vehicles exchange data with these access points at regular intervals to provide near real-time information. Our results show that even for a very simple topology (2 route alternatives), the improvements in average journey time can be up to 6%. With more complex cities the benefits will be far more significant. Note: This work was not funded by CMI/CII.
(53.4 KB) - added by dnc25
7 years ago.
(1.3 MB) - added by dnc25
7 years ago.
- IMG_2509small.2.JPG (26.3 KB) - added by dnc25 7 years ago.
(20.9 KB) - added by dnc25
7 years ago.
Photo of white van with sun behind mast.
(0.7 MB) - added by jjd27
4 years ago.