15.
M. Schreieck, M. Wiesche and H. Krcmar, “Design and Governance of Platform Ecosystems – Key Concepts and Issues for Future Research,” in Twenty-Fourth European Conference on Information Systems (ECIS), Istanbul, 2016.
16.
A. Hein, M. Schreieck, M. Wiesche and H. Krcmar, “Multiple-Case Analysis on Governance Mechanisms of Multi-Sided Platforms,” in Multikonferenz Wirtschaftsinformatik (MKWI), 2016.
17.
J. Manner, D. Nienaber, M. Schermann and H. Krcmar, “Six Principles for Governing Mobile Platforms,” in Wirtschaftsinformatik Proceedings 2013, 2013.
18.
D. Christin, “Privacy in mobile participatory sensing: Current trends and future challenges,” Journal of Systems and Software, vol. 116, pp. 57–68, 2016.Crossref
19.
V. Sucasas, G. Mantas, F. B. Saghezchi, A. Radwan and J. Rodriguez, “An autonomous privacy-preserving authentication scheme for intelligent transportation systems,” Computers & Security, pp. 193–205, 2016.
Footnotes
1This chapter is based on the following two publications: Schreieck, M.; Wiesche, M.; Krcmar, H. (2016). Modularization of Digital Services for Urban Transportation. Twenty‐second Americas Conference on Information Systems (AMCIS), San Diego and Pflügler C.; Schreieck, M.; Hernandez, G.; Wiesche, M. and Krcmar, H. (2016). A concept for the architecture of an open platform for modular mobility services in the smart city. International Scientific Conference on Mobility and Transport (mobil.TUM), Munich.
2We thank the German Federal Ministry for Economic Affairs and Energy for funding this research as part of the project 01MD15001D (ExCELL).
© Springer-Verlag GmbH Germany 2018
Claudia Linnhoff-Popien, Ralf Schneider and Michael Zaddach (eds.)Digital Marketplaces Unleashedhttps://doi.org/10.1007/978-3-662-49275-8_44
44. Safety Belt for Pedestrians
Klaus David1 and Hendrik Berndt1
(1)Universität Kassel, Kassel, Germany
Klaus David (Corresponding author)
Email: [email protected]
Hendrik Berndt
Email: [email protected]
44.1 Heads Up
“Heads up! Cross the street and then update Facebook” that’s how new street signs, installed to protect, seek pedestrians’ attention in Hayward, CA, USA. According to [1] distractions from mobile phones divert attention from traffic similar as for a person with 0.8 per mill blood alcohol. Obviously this fact might just be seen as a minor influence on the number of traffic fatalities in total. Latest numbers from 2/2016 give an overview on road safety evolution in the European Union. According to CARE (the EU road accidents database) 25,900 people were killed [2] in 2014 within the EU. Pedestrians and other unprotected road users are vulnerable and highest in number amongst most severe injured persons. Worldwide 22% of all fatalities are pedestrians [3]. Something more has to be done to protect them.
Today’s vulnerable road user (VRU) protection systems are mainly based on cameras, radar, infrared‐systems, microwaves or combination thereof. They are predominantly placed in vehicles and request a line of sight to pedestrians to be faultlessly recognized. This line of sight however is often not available.
Moreover, a dangerous situation has to be detected as early as possible to start accident avoidance measures. For that it would be desirable to have more information such as position, direction, velocity of movement from both pedestrians and cars involved, available or projected. There can even more relevant information to be exploited, such as closeness of the pedestrian to the curb, or his/her step onto the street level and even personalized information about the height of the person (child or grown up) the age (young or elderly both with different movement patterns at one’s disposal).
The presented solution allows for the pedestrian to play a decisive role in the prevention of accidents.
44.2 Communication Infrastructure
Our Pedestrian protection system assumes a mobile phone is available for the service user. Through optimal use of the phone’s full capabilities it becomes part of a communication infrastructure, which as an overall solution will provide alerts and protection to participants. Communications between pedestrians and affected cars is using existing technologies and does not necessarily request a specific infrastructure or standard. Several possibilities are at disposal, such as WLAN, cellular or device‐to‐device communications. Unquestionably latency has to be low enough to allow for several message exchanges during an average accident avoidance response time, available typically at 2 s, as analyzed in [4]. LTE with its ±50 ms round‐trip delay is providing a time cushion that allows for several message exchanges, and next mobile generation, 5G, is targeting even single digit millisecond delays.
Fig. 44.1 below depicts a potential communications infrastructure for the pedestrian’s safety belt.
Fig. 44.1ComTec 2016 – Communication infrastructure building blocks – 1 On Board Unit Inside Car, 2 Smartphone of pedestrian, 3 Ad‐hoc WLAN communications, 4 Cellular communications (UMTS/HPSA/LTE/…), 5 Potential server for scenario calculation (can also be done on Smartphone and On Board Unit only)
Amongst the different options for a communication infrastructure the smartphone based WiFiHonk system, introduced in [5] is noted. It uses WiFi Beacon Stuffing, a method to exchange information between WiFi devices, without establishing a connection, through the SSID or BSSID in the WiFi beacon header of a Wifi network. Wifi Beacon Stuffing can be used to propagate the latitude and longitude, the speed, and the direction of both VRUs and vehicles. These data can be transmitted every 100 ms and thus also be seemly as infrastructure unit of a pedestrian protection system. [6] introduces another communication infrastructure unit for cooperative applications between vehicle and pedestrians based on DSRC modules.
44.3 Context Filter
At the core of our protection system is an architectural building block called context filter. It interprets and transforms sensor data, gathered through the pedestrian’s mobile phone sensors into activity detection. Additionally it filters, fuses and evaluates available information from the pedestrian’s surrounding context. To protect privacy of pedestrians all information gathered would be anonymized trough a specific component inside the context filter, the anonymizer. (Legal groundwork for data protection has been elaborated on in [7]).
Our context filter consists out of the modules, depicted in Fig. 44.2.
Fig. 44.2ComTec 2016 – Context filtering and collision estimation
Once information is fused, filtered and evaluated a collision likelihood based on the situation is estimated and further actions are triggered if seen necessary. There are different possibilities to perform collision estimation e. g. at a central server, the onboard unit of the vehicle or the pedestrian’s mobile phone, as depicted in Fig. 44.1.
Each context filter module follows its own mission as exemplified by the following instances:
Position Module
Positioning is based on a global navigation system such as Global Positioning System (GPS). GPS modules in Smartphones today however with a precision about 5 to 8 m and a sampling frequency of 1 Hz do not fulfill the necessary requirements for a safe pedestrian protection system. Our solution combines GPS with information from acceleration sensor and gyroscope for better precision and faster recognition of a position change. For the motion recognition, the data of the acceleration sensor and the gyroscope in the x‑, y‐ and z‐axis are captured with a sample rate of 40 Hz. Motion recognition is combined with walking speed estimation to develop a dead reckoning algorithm for pedestrians that takes their current movement state into account. Details about the algorithm can be found in [8].
Direction Module
This module gathers the movement of a pedestrian relative to the street utilizing th
e built‐in magnetometer sensor of a smartphone. Results are important for verifying the likelihood of a collision with a vehicle. All pedestrians moving away from the street are not endangered; others have to be further observed, taking the information about the position, direction, speed and acceleration of the respective cars ‐provided by the vehicle detection module – into account. Attainable direction accuracy of the digital compass has been described through experiments in [9] and assessed as being suitable. However special attention shall be given to the impact of magnetic diversions since they influence the accuracy of the magnetometer (e. g. through parking cars). Our algorithm developed allows for necessary corrections and to compensate for any magnetic deviation, simply by subtracting the assessed deviation from the magnetometer sensor data in order to achieve a more precise movement direction detection of pedestrians.
Curb Detection Module
Our method developed for curb detection has shown, that with smartphone sensor data from accelerometer and gyroscope and combined with an appropriate classifier, it is possible for the system to recognize when the pedestrian steps down onto the street level. This stepping on the road detection is one of the most relevant indications for higher collision likelihood. Due to the short duration of “curb crossing”, only 0.56 s on average, sliding window parameters are chosen to support its detection. In [10] a large number of experiments have been conducted to evaluate achievable results for stepping down from the curb recognition, taking into account several parameters such as different curb heights and different walking speeds. The context filter fuses this information about “curb crossing” into the situation assessment as a specific trigger for collision risk evaluation.
Complimentary approaches to solve the street crossing detections for pedestrians exist. In [11] street crossing detection relies only on GPS data. The street crossing algorithm checks if the path of a pedestrian will cross the street by extrapolating the user’s path from its position using the bearing values from the GPS data over a fixed distance. In the evaluation it was found that the algorithm was able to detect 85% of all street crossings in a suburban and 78% in an urban environment at a maximum of 5 s after the person actually stepped on the street. The authors of [11] conclude that GPS only “does not serve well the fine‐grained positioning needs of pedestrian safety applications in dense urban environments”.
Motion Profile Module
The context filter offers to create a pedestrian specific motion profile. It provides important information that concern movement behavior patterns, maximum speed capabilities, age and height and additional personalized information about the pedestrian involved to be utilized for an even more efficient protection procedure. Of course under strict observance of privacy and data protection. Guidelines and first approaches have beenpublished in [12].
Finally – after combining and filtering data available from all context filter modules – a collision estimation module computes the likelihood of an accident and if appropriate starts preset warning procedures, which can be acoustic, optic or haptic or any combination thereof for user alerts, as depicted in Fig. 44.3. Within the collision avoidance scenarios “missed alarms” and “false alarms” are important performance indicators, they have been sampled for each of the context filter’s modules and have to be continuously evaluated for improving the overall accuracy of collision estimation results.
Fig. 44.3ComTec 2016 – Safety Belt for pedestrian overall solution
44.4 Business Case Scenarios
The proposed “safety belt for pedestrian” solution provides a compelling approach to reduce an ever‐increasing number of traffic casualties from vulnerable road users. Its market potential builds upon the high acceptance of communities and the society as a whole to overcome traffic fatalities and to reduce the related expenses. With that in mind it is envisioned that many stakeholders, from health insurance companies to elderly care centers, from safe community activists to OEM manufacturer will participate in a valuable set of business cases. The safety belt for pedestrian thus can foster business liaisons for all interested parties and establish a sustainable business organization for market coverage and growth.
A first use case could be based on low monthly fee for participants, who use the service. Other use cases foresee free services from e. g. infrastructure provider or apply a “pay per use”, where service is only charged for if a warning has been issued to the pedestrian.
44.5 Conclusion
The paper presents a protection system for vulnerable road users, with minimal investments and high efficiency. It is distinct from existing road user protection efforts in particular through: No line of site as a prerequisite for pedestrian’s detection
The proposed solution works weather independent
Time advantage, through early recognition of a developing dangerous situation, which will allow for earlier triggering of alerts and actions.
Still a risk for the best possible impact of the overall system remains, since the complexity for filtering and analyzing the situational data at hand is rather large and naturally very time‐critical. Therefore the efficiency of the core element “context‐filter” is of outmost significance. However a huge market acceptance for the proposed pedestrian protection system can be expected since it will interest many different players and all of the in public health care involved parties to participate in sustainable business cases, to overcome traffic fatalities and to considerably reduce related expenses.
References
1.
S. Huber, “Kopf hoch, bitte!,” ADAC Motoerwelt Magazin, pp. 34–35, 07 2015.
2.
[Online]. Available: http://ec.europa.eu/transport/road_safety/pdf/observatory/historical_evol.pdf. [Accessed 31 08 2016].
3.
World Health Organization, Global status report on road safety 2015: supporting a decade of action, Geneva, Switzerland: WHO, 2015.
4.
A. Flach und K. David, “A Physical Analysis of an Accident Scenario vetween Cars and Pedestrians,” in IEEE 70th Vehicular Technology Conference (VTC 2009-Fall), Anchorage, AK, USA, 2009.
5.
K. Dhondge, S. Song, B.-Y. Choi und H. Park, “WiFiHonk: Smartphone-Based Beacon Sutffed WiFe Car2X-Communication System for Vulnerable Road User Safety,” in IEEE Vehicular Technology Conference (VTC 2014-Spring), Seoul, South Korea, 2014.
6.
A. Tahmasbi-Sarvestani, H. Kazemi, Y. Fallah, M. Naserian und A. Lewis, “System Architecture for Cooperative Vehicle-Pedestrian Safety Applications Using DSRC Communication,” in SAE 2015 World Congress & Exhibition, Detroit, MI, USA, 2015.Crossref
7.
T. Schulz, A. Roßnagel und K. David, Datenschutz bei kommunizierenden Assistenzsystemen, Bd. 11, Für Datenschutz ZD, 2012, pp. 510–515.
8.
S. Engel, K. David, D. Warkow und M. Holzknecht, “Car2Pedestrian Positioning: Methods for Improving GPS Positioning in Radio-Base VRU Protection Systems,” in 6. Tagung Fahrerassistenzsysteme, Munich, Germany, 2013.
9.
A. Memon, S. Lau und K. David, “Investigation and Compensation of the Magnetic Deviation on a Magnetometer of a Smartphone Caused by a Vehicle,” in IEEE 78th Vehicular Technology Conference (VTC 2013-Fall), Las Vegas, NV, USA, 2013.Crossref
10.
A. Jahn, K. David und S. Engel, “5G/LTE based protection of vulnerable road users: Detection of crossing a curb,” in IEEE 82nd Vehicular Technology Conf. (VTC2015-Fall), Boston, MA, USA, 2015.
11.
S. Jain, C. Borgiattino, Y. Ren, M. Gruteser und Y. Chen, “On the limits of position-based pedestrian risk awareness,” in Proc. Workshop Mobile Augmented Reality and Robitiy Technology-based Systems (MARS-2014), Bretton Woods, NH, USA, 2014.
12.
O. Coutland, S. Lau und K. David, “Determining the behavior of mobi
le location-aware services based on user action history,” in GIITG KuVS Fachgespr. Selbstorganisierende Adapt. Kontextsensitive Verteilte Syst. SAKS, Kassel, Germany, 2006.
Further Reading
13.
K. David und A. Flach, “CAR-2-X and Pedestrian Safety: Innovative Collision Avoidance System,” in IEEE Veh. Technol. Mag., 2010.
© Springer-Verlag GmbH Germany 2018
Claudia Linnhoff-Popien, Ralf Schneider and Michael Zaddach (eds.)Digital Marketplaces Unleashedhttps://doi.org/10.1007/978-3-662-49275-8_45
45. The Impact of Indoor Navigation Systems for Public Malls – a Comprehensive Overview –
Karsten Weronek1
(1)Frankfurt University of Applied Sciences, Frankfurt on the Main, Germany
Karsten Weronek
Email: [email protected]
Abstract
This paper introduces the most favored technologies for Indoor Positioning Systems and the features of Indoor Navigation Systems for malls. These systems guide the customer through a building on his way to a desired place. On the other hand it makes possible to analyze the customer’s journey. Combined with retail business apps on the customers’ smartphones promises to generate revenue and customer loyalty. An intellectual stakeholder analysis reveals the potential benefits and issues for the future. However, the technology is not mature yet, business integration is complex and the implementation and operation of such a system has to mitigate various risks to be able to reach its return of investment.
Digital Marketplaces Unleashed Page 68