Designing a Smart e-Bike Eco-System 

Get Complete Project Material File(s) Now! »

Smart Mobility and Sensing: case studies based on a Bike Information Gathering Architecture

This chapter discusses PUMA, a Personal Urban Mobility Assistant designed to overcome classic limited set of means of transports o ered by mapping services. PUMA aims to let the user add di erent factors of personalization, such as sustain-ability, street and personal safety, wellness and health. PUMA was also the prelude to the Canarin project, described in chapter 3. In this scenario, the sensor was equipped on a smart bike, as a mean to collect data about the urban environment. The Bike Information Gathering Architecture (BIGA) is also described herein and it may be seen as a preliminary deploy attempt to the Canarin’s Cloud Architecture, designed only at a later time.

Introduction

Smart mobility is playing a strategic role in our daily life in the urban scenario, taking into account that most people live in cities [16, 17]. Thanks to the wide di usion of mobile devices, several services and applications based on the geograph-ical position of users are now available [18], which support citizens while they move across the urban environment [19]. In this context, personalization can be a key factor, enabling independence of citizens, despite some speci c conditions (i.e. dis-abilities) [20] or means of transport [21]. Existing mapping services (e.g. Google Maps) and travel planners (e.g. OpenTripPlanner, Graphhopper) provide multi-modal paths computed by the same algorithms, on the basis of the same elements: time, distance, and a limited set of means of transports (e.g. cars, public buses and metros, feet). However, it is not possible to add di erent factors of personalization, in terms of sustainability, street and personal safety, wellness and health, mood and satisfaction, accessibility. In this context, our idea is to design a system which acts as a Personal Urban Mobility Assistant (PUMA), supporting citizens in:
• collecting di erent information about the urban environment (by means of crowdsourcing and crowdsensing activities) in terms of: pollution, tra c, safety, health and tness, etc.;
• exploiting gathered data, proving multimodal and multipreference paths in the urban environment.
In order to provide multimodal and multipreference paths in urban environments, a detailed mapping of all the elements that a ect these factors (e.g., data about pollution, urban barriers and facilities, street lights, data about car accidents, data about crimes, etc) is needed. Moreover, given out this information, it is necessary a system that lets each user customize and modulate the route computation on the basis of his/her own needs, instead of using the same algorithm for all.
As regards the mapping, our approach is based on an open and participatory sensing and mapping system, with low cost sensors, which exploit users’ devices too, in a common and shared data repository. We would take advantages by the poten-tialities of cloud architecture to create a modular open and crowdsourced system. In our work, we also tackled the following challenges:
• To introduce an innovative users’ approach towards mobility choices that matches all impact factors for transportation, driving di erent transportation services, from single to shared, going next to the common existing booking sys-tems, o ering a social environment to share experiences and information on sustainable mobility and participate to challenges, info on tra c, lane condi-tion and pollution [22], with the aim of supporting and improving eco-driving and sustainable behaviours [23].
• To develop a smart urban approach to mobility based upon way of booking transport systems that also take into account the carbon footprint [24].
• Integration of sustainable eets with public transportation (i.e. buses and train) with the possibility to buy tickets by smart payment systems too.
• Data storage and data management (integration from di erent data sources: from public transportation and route conditions to air pollution obtained by sensors installed on bicycles or other vehicles [25], integration with tra c info).
In this chapter, we focus on a speci c means of transports: bikes [26]. Bikes can be equipped with di erent kinds of sensors and can be connected each other, so as to create a speci c vehicular network, integrated with the urban infrastructure [27, 28] and networks [29] thanks to a cloud architecture. The chapter also describes the system architecture and details personas and related scenarios, showing how it can be exploited by di erent users, with di erent needs and preferences, applying an altruistic IoT approach [30].
The remainder of the chapter is structured as follows. Section II describes the system architecture. Section III de nes some personas and Section IV presents some use scenarios. Finally, Section V concludes the chapter highlighting some nal remarks and future work.

Cloud Architecture

In this section, we introduce our system architecture, speci cally thought for bicycle vehicles, named BIGA (Bike Information Gathering Architecture), shown in Figure 2.1. Our previous work [21] focused mainly on the adoption and implementation of a speci c software engineering model that envisioned every component of a mobility application as a service; the reference model was based on microservices, therefore the SMAll architecture was tailored at proposing the de nition of an open and standard interface for service access. Instead, BIGA architecture describes at high level the physical and software architecture that might be put in place to provide smart bicycle services. This means that BIGA might be adopted to host and provide the implementation proposed in the SMAll project. Bikes are equipped with sensing devices capable of gathering di erent kinds of information (data) not only from the environment (e.g., air pollution [14]), but also from the bike itself (e.g., traveled kilometers via odometer). Once (periodically) collected, such information are sent to an entity devoted either to provide connectivity or forwarding data to the cloud via Internet; this entity might be an infrastructure located along the road, such as for example a roadside unit (RSU), or a speci c gateway. Cloud infrastructure is where data are processed, stored and made available for being consumed by multiple users.
The idea is to allow users, who are interested in gathering information, to per-sonalize a plan for a given path, depending on their daily habits or needs. The cloud hosts the software that provides path customization and other useful services, but targeted for di erent uses, as described in the next section. This implies that di er-ent applications might require di erent ways of data collection. Therefore, multiple users can source information by accessing, for example, a web application in order to properly plan in advance their path, or use a mobile application on the smartphone not only for a priori decision, but for real time consultation as well. This means that data can be consumed prior and/or during the journey. At the end of the journey, users can decide to share their experience, i.e., share \collected information » along the path; this would allow to enrich databases with new information, thus resulting in improved bikers experience that can bene t of feedbacks coming from the com-munity. Therefore, with our approach, users are both consumer and producer, so that data are both gathered and disseminated from/in the community.
The vision is that the Municipal District of Bologna (MDB) might act as service provider, i.e., providing to citizenship such \smart bicycles sharing » system. Smart bicycles would be equipped with devices targeted for the di erent applications (e.g., air pollution monitoring, tness monitoring, personal safety and carbon footprint). MDB would then rent cloud resources at an infrastructure provider, whose goal would be to provide computational, network and storage resources to have service in place, besides the mobile and web application needed to interact with the service by remote users. Making the cloud hosting the applications, and making these applications available to customers, allow our \Bike as a Service » to fall under the hat of Software as a Service (SaaS). Indeed, cloud approach adoption brings several bene ts:
• MDB has no need to install and run applications on their own computers, resulting in less expense in terms of buying new hardware, infrastructure pro-visioning and consequent maintenance.
• Other emergent paradigms might be put in place on need: Network Func-tion Virtualization (NFV) and Software De ned Networking (SDN) might be adopted in synergy with cloud in order to provide exible, programmable and cost e ective solutions [31]. For example, software applications might run on a Virtual Machines (VMs) interconnected by a virtual network [32]; NFV would help in delivering services as virtual functions, while SDN would help in exibly managing the (virtual) network [33].
• Cloud approach also calls for service on-demand model, that is, virtual func-tions could be instantiated, removed or migrated across the network without the need of deploying new hardware.

Personas

In this section, we present three personas designed with the aim of de ning scenarios exploiting di erent characteristics of our system.
Wei
Wei is a Chinese-American visiting Scholar, working to a joint research in Bologna for three months. His research interests are related to climate change and he is very committed in reducing his individual carbon footprint. These days in Bologna are mainly devoted to work and complete all the scheduled research tasks and experi-ments. In his free time, he likes to go around the city centre and explore the old town of Bologna (Figure 2.2).
Wei works at the University of California, San Diego since 2007. He is in Bologna now for a joint research on climate change in the Department of Biological, Geolog-ical and Environmental Sciences of the Bologna University. Wei’s family is based in S. Diego, where his wife Xiu Ying and his two children (Sean and May) live. They keep in touch with a daily call and Wei sends them a lot of pictures taken wandering around the city while he rides his bike. While in Bologna, Wei lives in the University guest quarters (Residenza di San Giovanni in Monte) located in a prominent monumental complex belonging to the University of Bologna, in the old town centre of Bologna. Wei uses a good trekking bike, loaned by a colleague from the Department for his stay in Bologna. He sporadically uses bus and other public means to reach destination that are too far from the city centre to be reached by bicycle. This responds both to Wei commitment to use sustainable transport and to his travel needs.
Sven
Sven is a Swedish Erasmus student, living in Bologna for 6 months to complete his master thesis. He is vegan and he is obsessed with tness. Stay in very good shape, eat vegan and whole food, have a generally healthy life style are very important goals to Sven. During his stay in Bologna, Sven goes to the Department to work to his thesis, to the gym to do his workout, without forgetting to go around with friends, having fun and enjoying the city life as all students do.
He is studying Film Directing at the Faculty of Fine, Applied and Performing Arts at the University of Gothenburg and he is completing his thesis on the Kill Bill movie series by Quentin Tarantino, working in the Department of Arts of the Bologna University. Sven is single. His family of origin lives in Hastevik, a small town near Gothenburg. They keep in touch on a weekly basis with a conference call. While in Bologna, Sven lives in a shared apartment in a neighbourhood outside the city centre. He decided for this location to share the room with his friend Hugo, who is taking his master degree in Economics and Finance in Bologna. The apartment is quite near to the Business and Economics School, where Hugo studies, but pretty far (about 4 Km) from the Arts Department. Sven bought a cheap used mountain bike from another Erasmus student leaving Bologna few days after his arrival. He uses a mix of bike and bus to move around the city, depending from weather, time of the day, distance of the destination, but the bicycle is the most used mean to go to the Department and to the Gym on a daily basis because it represents an opportunity to do more workout and also to save money.
Elena
Elena works full time at the University of Bologna. She was born in Bologna and she grow up in its city centre. Since her husband works in a nearby city, she is in charge of managing their son Tommaso (Tommy) and travel with him to and from school, or to and from her parents’ house. While Elena prefers to use the car to reach points of interest outside the centre, when she goes downtown (to work, to Tommy’s school and to her parents’ house), she prefers to leave the car at home. Elena likes to ride the bike, but she is very worried for Tommy, by the safeness of the travel, weather issues and pollution that can be dangerous, especially during the winter. Elena works for the University of Bologna since 2008. She works in the International Desk, providing information to students wishing to enroll at the University. She works since 8.30 AM to 4.30 PM for 5 days a week, having a fast lunch in the o ce nearby. Elena is married with Alberto since 2010 and they have a son, who is 4 years old. Tommy goes to a primary school (Betti Giaccaglia Plesso 2) located in the Montagnola Park, immediately near the Bologna Station. Elena got the option to enroll Tommy in this school because her father and mother live nearby and they are used to pick up Tommy from school every day at 4 PM. Tommy waits for Elena in grandparents’ house, located in via Mascarella, for about an hour, to come back home with her. Elena lives in Bolognina, a neighbourhood outside the city centre, quite near to the train station. His husband, Alberto, works in Cesena and this location was chosen mainly to meet his need to easily reach the station. The place is not far from Tommy’s School (about 1 Km) and from Elena workplace (about 2 km), hence Elena uses a new red city bike, fully equipped with lights and re ectors, to enhance safeness of the travel, and with a baby seat on the back.

Travel scenarios

Travel scenarios related to the personas introduced in the previous section are de-scribed in the following subsections.
Wei
Wei is going to the weekly meeting of the research team, to reschedule some late experiments. It’s a foggy day, but despite cold and humid, Wei is happy to take the bike. The meeting will be at 9.00 AM, Wei is leaving the University guest quarters early, so as to have a sweet breakfast in a bar near the Department without being in a hurry. Having more time than what is strictly required to reach his destination, he decided to enjoy the ride and cross the city mainly passing through restricted tra c zones.
In Figure 2.5, there are two di erent paths between Wei’s starting point (Via de Chiari) and destination (Via Selmi). On the left side, there is the default and shortest path proposed for bikes by GraphHopper. On the right side, there is the personalized one based on the user’s preferences, computed by our PUMA. This latter avoids one of the most congested and polluted roads of Bologna.
Sven
This morning Sven is going the Cineteca di Bologna to study some sources and will reach at noon his master thesis supervisor at the Department. The weather is not perfect, it is partially cloudy, but Sven prefers to use the bike because he will not have enough time for the Gym. So Sven prefers a longer path so as to do a good workout. Our PUMA proposes a longer path, a path through di erent green areas and some slopes, as shown in Figure 2.6.
Elena
Elena is going to work, she will leave Tommy at the school on her way. The weather is very good, 25 C, a perfect spring day with a perfect temperature. She is leaving in at 7.45 AM, just in time to stop at the School, say bye to Tommy and go to work in schedule without being in a hurry. She decides to go safe using the available bicycle lanes and to select a route through parks and green areas to enjoy the spring weather and avoid a large exposure to pollutant. Our system proposes a path that does not cover Via Irnerio (as shown in Figure 2.7), a route not safe for cyclist because of the tra c, that includes cars and buses, and not clean, due to the pollution produced by these means of transport.

Table of contents :

1 Introduction 
2 Smart Mobility and Sensing: case studies based on a Bike Information Gathering Architecture 
2.1 Introduction
2.2 Cloud Architecture
2.3 Personas
2.4 Travel scenarios
2.5 Conclusions
3 Canarin II: Designing a Smart e-Bike Eco-System 
3.1 Introduction
3.2 Background and Related Work
3.3 Design Issues
3.3.1 Health
3.3.2 Wellness
3.3.3 Safeness
3.4 Our Prototype Architecture
3.5 Prototype Presentation
3.6 Conclusions
4 Canarin Nano: a wearable pollution monitor for the allergic rhinitis and sleep quality case study 
4.1 The context: the POLLAR clinical study
4.2 Design
4.2.1 Particle Electron
4.2.2 Remote control
4.3 Implementation
4.3.1 Server side and frontend
4.3.2 Hardware design
5 Monitoring cultural heritage buildings via low-cost edge computing/sensing platforms: the Biblioteca Joanina de Coimbra case study 
5.1 Introduction
5.2 The spatial context
5.3 Monitoring Campaign
5.4 Architecture
5.5 Preliminary data analysis
5.6 2020 Updates
5.7 Final remarks
6 On Assessing the Accuracy of Air Pollution Models Exploiting a Strategic Sensors Deployment 
6.1 Introduction
6.2 Air pollution models assessment
6.3 The spatial context
6.4 The sensors station
6.5 The sensors deployment
6.6 Conclusion and future works
7 Over The Air EV monitoring and analytics: the Nissan Leaf case 
7.1 The CAN Protocol
7.1.1 OBD-II: on-board diagnostic
7.2 Energy consumption compaign
7.3 CAN variables from the Nissan Leaf 2018
7.3.1 Vehicle Control Module (VCM)
7.3.2 Body Control Module (BCM)
7.3.3 Li-Ion Battery Controller (LBC)
7.3.4 Antilock Braking System (ABS)
7.3.5 Traction Motor Inverter (TMI)
7.3.6 Meter
7.3.7 Heating, Ventilation and Air Conditioning (HVAC)
7.4 Evaluation
7.4.1 Experimental setup
7.4.2 Data analysis
8 X-Fi: Revisiting WiFi Ooading in the Wild for V2I Applications
8.1 Introduction
8.2 Related Work
8.3 System Overview
8.3.1 X-Fi: connecting a moving vehicle to commercial WiFi hotspots
8.3.2 X-Perf : measuring intermittent connectivity
8.4 Measurement Methodology
8.4.1 Experimental Setup
8.4.2 Connectivity Metrics
8.4.3 Network Performance Metrics
8.5 Data Collection
8.5.1 Dataset Summary
8.6 Micro Benchmarks
8.6.1 WiFi association time
8.6.2 Time until IP address acquisition
8.6.3 Link-layer connectivity duration
8.6.4 IP connectivity duration
8.6.5 Connectivity Holes
8.6.6 IP Address Roaming Support
8.6.7 Impact of Speed, ISS, and frequency
8.7 Network Performance Analysis
8.7.1 Time until the rst TCP ACK
8.7.2 TCP connectivity duration
8.7.3 TCP connectivity holes
8.7.4 Average TCP goodput
8.7.5 Data volume
8.8 Discussion and Final Remarks
8.9 X-Fi implementation details
8.9.1 X-Fi orchestrator role
8.10 X-Perf Implementation details
9 C-Continuum 
9.1 Background
9.2 System design
9.2.1 Overview
9.2.2 Architecture
9.2.3 Naming
9.2.4 Caching
9.2.5 Security
10 Conclusions

GET THE COMPLETE PROJECT