FullStop: Tracking unsafe stopping behaviour of buses
01 Jan 2018-pp 65-72
TL;DR: FullStop, a smartphone-based system to detect safety risks arising from bus stopping behaviour, is presented, which is based on the view obtained from looking out to the front of the vehicle using the camera of a smartphone that is mounted on the front windshield.
Abstract: Road safety is a critical issue the world-over, and the problem is particularly acute in developing countries, where the combination of crowding, inadequate roads, and driver indiscipline serves up a deadly cocktail. We believe that mobile devices can play a positive role in this context by detecting dangerous conditions and providing feedback to enable timely redressal of potential dangers. This paper focuses on a specific problem that is responsible for many accidents in developing countries: the stopping behaviour of buses especially in the vicinity of bus stops. For instance, buses could arrive at a bus stop but continue rolling forward instead of coming to a complete halt, or could stop some distance away from the bus stop, possibly even in the middle of a busy road. Each of these behaviours can result in injury or worse to people waiting at a bus stop as well as to passengers boarding or alighting from buses. We present FullStop, a smartphone-based system to detect safety risks arising from bus stopping behaviour, as described above. We show that the GPS and inertial sensors are unable to perform the fine-grained detection needed, by themselves. Therefore, FullStop is based on the view obtained from looking out to the front of the vehicle using the camera of a smartphone that is mounted on the front windshield. Using optical flow vectors, with several refinements, FullStop running on a smartphone is able to effectively detect safety-related situations such as a rolling stop or stopping at a location that is displaced laterally relative to the designated bus stop.
Citations
More filters
TL;DR: A novel, cost-effective Internet of Things (IoT) architecture is introduced that facilitates the realization of a robust and dynamic computational core in assessing the safety of a road network and its elements and a new, meaningful, and scalable metric for assessing road safety.
Abstract: The Safe System (SS) approach to road safety emphasizes safety-by-design through ensuring safe vehicles, road networks, and road users. With a strong motivation from the World Health Organization (WHO), this approach is increasingly adopted worldwide. Considerations in SS, however, are made for the medium-to-long term. Our interest in this work is to complement the approach with a short-to-medium term dynamic assessment of road safety. Toward this end, we introduce a novel, cost-effective Internet of Things (IoT) architecture that facilitates the realization of a robust and dynamic computational core in assessing the safety of a road network and its elements. In doing so, we introduce a new, meaningful, and scalable metric for assessing road safety. We also showcase the use of machine learning in the design of the metric computation core through a novel application of Hidden Markov Models (HMMs). Finally, the impact of the proposed architecture is demonstrated through an application to safety-based route planning.
15 citations
Cites methods from "FullStop: Tracking unsafe stopping ..."
...Smartphone sensing is also utilized in [38] using both accelerometers and cameras to identify unsafe stopping behavior in busses....
[...]
04 Jan 2022
TL;DR: In this article , a leader-based hierarchical clustering algorithm was proposed to reveal public bus-stops from the GPS traces with 92% and 95% accuracy, which can help transport policymakers make a timely decision for curbing stop irregularity to a large extent.
Abstract: Municipal authorities in the suburban cities of many developing countries need to deal with rapid unplanned urbanization, huge population bursts, and policy planning under severe budget shortfalls. We have studied the clear clutter and disorder in public bus-stop patterns that prevail upon the city transportation system in such cities. This paper develops an auto-tuned mechanism for identifying the bus-stops and characterizing their Spatiotemporal features from the GPS trails obtained through spatial crowdsensing. Existing systems and algorithms cannot be merely replicated in this context because of the kind of heterogeneity and chaotic situation prevalent in the transportation system. In this paper, we analyze GPS traces of more than 6000km (approx.) bus navigation, which has been collected for two years (2014–2016) via automated war-driving through customized hardware, for bus routes at a suburban city, Durgapur, in India. We develop a novel leader-based hierarchical clustering algorithm to reveal public bus-stops from the GPS trails with 92% and 95% accuracy. The proposed approach can help transport policymakers make a timely decision for curbing stop irregularity to a large extent.
2 citations
01 Dec 2018
TL;DR: A novel application of Hidden Markov Models (HMMs) at the core of road safety assessment is introduced and the use of the dynamic assessment in safety- based route planning is demonstrated.
Abstract: The Safe System approach to road safety emphasizes safety-by-design through ensuring safe vehicles, road networks, and road users. With a strong WHO/UN motivation, the approach is increasingly adopted worldwide. Considerations in Safe System, however, are largely made for the medium-to-long term. Our interest in this work is complement the approach with a short-to-medium term dynamic assessment of road safety. Towards this end, we introduce a novel application of Hidden Markov Models (HMMs) at the core of road safety assessment. We also demonstrate the use of the dynamic assessment in safety- based route planning.
Cites background from "FullStop: Tracking unsafe stopping ..."
...This includes, for instance, the work in [9] utilizing smartphones sensors (accelerometers and cameras) to identify unsafe stopping behavior in busses....
[...]
References
More filters
28 Jun 2011
TL;DR: SignalGuru as discussed by the authors is a software service that relies solely on collaborating windshield-mounted mobile phones to provide information about the schedule of traffic signals and enable a set of novel driver-assistance applications.
Abstract: While traffic signals allow competing flows of traffic to safely cross busy intersections, they inevitably enforce a stop-and-go movement pattern. This stop-and-go movement pattern increases fuel consumption by 17%, CO2 emissions by 15%, and reduces vehicle flow aggravating congestion and driver frustration.The stop-and-go movement pattern can be alleviated by utilizing information about the future schedule of the traffic signals ahead. Based on when the signal ahead will turn green, onboard computational devices (e.g., smartphones) can advice the drivers on the optimal speed they should maintain so that they can cruise through an intersection and avoid coming to complete halt. Alternatively, efficient detours may be suggested to the drivers to avoid waiting for a long time at a red light.Our MobiSys'11 paper proposes SignalGuru, a novel software service that relies solely on collaborating windshield-mounted mobile phones to provide information about the schedule of traffic signals and enable a set of novel driver-assistance applications.The SignalGuru service consists of four main modules: First, video frames are captured with the mobile phone cameras and processed to detect the color (status) transitions (e.g., RED to GREEN) of the traffic signal ahead (detection module). Then, information across multiple consecutive frames is used to filter away erroneous traffic signal transition detections (transition filtering module). Third, SignalGuru-enabled phones collaborate by sharing their databases of detected traffic signal transitions with other phones within communication range (collaboration module). Finally, the merged data- base of traffic signal transitions is fed into a customized machine learning-based model to predict the future schedule of the traffic signal ahead (prediction module).Our demo will present a SignalGuru system, in which two SignalGuru-enabled iPhone 4 devices (iPhones A and B) will be collaborating to predict the schedule of the two pairs of mock traffic signals of the two intersecting roads of an intersection. The pictures of the intersection (taken with actual windshield-mounted iPhone devices) will be printed on posterboards A and B. The camera of SignalGuru-enabled iPhone A will be looking at posterboard A and the camera of SignalGuru-enabled iPhone B at posterboard B. While the iPhone B device will be fixed, the iPhone A device will be mounted on a lego vehicle so that it can move closer to or farther away from posterboard A. The mock traffic signals will be implemented with electronic displays (iPhone/iPad screen) and will be switching to red/green/ yellow based on a defined schedule.For the demo, we will be be bringing all the required electronic equipment and the two posterboards. However, we will also need two holders for the posterboards and two small tables for Signal-Guru-enabled devices A and B to stand.Demo attendants will be able to test the real time operation of SignalGuru and interact with it. More specifically, demo attendants will be able to see how the detection window gets dynamically adapted as they change the device's orientation and/or its distance from the traffic signal, and how SignalGuru's detection and filtering modules react to real-world events like fully or partially occluded traffic signals and false positive/negative signal detections.We believe that our SignalGuru demo will complement nicely our paper presentation as it will offer a good opportunity to discuss with conference participants about our system, demonstrate to them the challenges that our system poses as well as how it tackles them.
402 citations
TL;DR: This paper presents a comprehensive survey of the recent advances in activity recognition with smartphones' sensors, starting with the basic concepts such as sensors, activity types, etc and reviewing the core data mining techniques behind the main stream activity recognition algorithms.
335 citations
"FullStop: Tracking unsafe stopping ..." refers background in this paper
...But accelerometer usage has been explored for activity recognition in the past [22], [23]....
[...]
25 Jun 2013
TL;DR: CarSafe is the first dual-camera sensing app for smartphones and represents a new disruptive technology because it provides similar advanced safety features otherwise only found in expensive top-end cars.
Abstract: We present CarSafe, a new driver safety app for Android phones that detects and alerts drivers to dangerous driving conditions and behavior. It uses computer vision and machine learning algorithms on the phone to monitor and detect whether the driver is tired or distracted using the front-facing camera while at the same time tracking road conditions using the rear-facing camera. Today's smartphones do not, however, have the capability to process video streams from both the front and rear cameras simultaneously. In response, CarSafe uses acontext-aware algorithm that switches between the two cameras while processing the data in real-time with the goal of minimizing missed events inside (e.g., drowsy driving) and outside of the car (e.g., tailgating). Camera switching means that CarSafe technically has a "blind spot" in the front or rear at any given time. To address this, CarSafe uses other embedded sensors on the phone (i.e., inertial sensors) to generate soft hints regarding potential blind spot dangers. We present the design and implementation of CarSafe and discuss its evaluation using results from a 12-driver field trial. Results from the CarSafe deployment are promising -- CarSafe can infer a common set of dangerous driving behaviors and road conditions with an overall precision and recall of 83% and 75%, respectively. CarSafe is the first dual-camera sensing app for smartphones and represents a new disruptive technology because it provides similar advanced safety features otherwise only found in expensive top-end cars.
189 citations
TL;DR: A survey of smartphone-based insurance telematics is presented, including definitions; Figure-of-Merits (FoMs), describing the behavior of the driver and the characteristics of the trip; and risk profiling of theDriver based on different sets of FoMs, characterized in terms of Accuracy, Integrity, Availability, and Continuity of Service.
Abstract: Smartphone-based insurance telematics or usage based insurance is a disruptive technology which relies on insurance premiums that reflect the risk profile of the driver; measured via smartphones with appropriate installed software. A survey of smartphone-based insurance telematics is presented, including definitions; Figure-of-Merits (FoMs), describing the behavior of the driver and the characteristics of the trip; and risk profiling of the driver based on different sets of FoMs. The data quality provided by the smartphone is characterized in terms of Accuracy, Integrity, Availability, and Continuity of Service. The quality of the smartphone data is further compared with the quality of data from traditional in-car mounted devices for insurance telematics, revealing the obstacles that have to be combated for a successful smartphone-based installation, which are the poor integrity and low availability. Simply speaking, the reliability is lacking considering the smartphone measurements. Integrity enhancement of smartphone data is illustrated by both second-by-second lowlevel signal processing to combat outliers and perform integrity monitoring, and by trip-based map-matching for robustification of the recorded trip data. A plurality of FoMs are described, analyzed and categorized, including events and properties like harsh braking, speeding, and location. The categorization of the FoMs in terms of Observability, Stationarity, Driver influence, and Actuarial relevance are tools for robust risk profiling of the driver and the trip. Proper driver feedback is briefly discussed, and rule-of-thumbs for feedback design are included. The work is supported by experimental validation, statistical analysis, and experiences from a recent insurance telematics pilot run in Sweden.
139 citations
"FullStop: Tracking unsafe stopping ..." refers background in this paper
...Finally, we also note that there is growing interest in insurance telematics using smartphones and/or other dedicated sensors [15], [16]....
[...]
18 May 2015
TL;DR: The extensive evaluation results show that V-Sense is accurate in determining and differentiating various steering maneuvers, and is thus useful for a wide range of safety-assistance applications without additional sensors or infrastructure.
Abstract: Detecting how a vehicle is steered and then alarming drivers in real time is of utmost importance to the vehicle and the driver's safety, since fatal accidents are often caused by dan- gerous steering. Existing solutions for detecting dangerous maneuvers are implemented either in only high-end vehicles or on smartphones as mobile applications. However, most of them rely on the use of cameras, the performance of which is seriously constrained by their high visibility requirement. Moreover, such an over/sole-reliance on the use of cameras can be a distraction to the driver. To alleviate these problems, we develop a vehicle steering detection middleware called V-Sense which can run on commodity smartphones without additional sensors or infrastructure support. Instead of using cameras, the core of V-Sense/ senses a vehicle's steering by only utilizing non-vision sensors on the smartphone. We design and evaluate algorithms for detecting and differentiating various vehicle maneuvers, including lane-changes, turns, and driving on curvy roads. Since V-Sense does not rely on use of cameras, its detection of vehicle steering is not affected by the (in)visibility of road objects or other vehicles. We first detail the design, implementation and evaluation of V-Sense and then demonstrate its practicality with two prevalent use cases: camera-free steering detection and fine-grained lane guidance. Our extensive evaluation results show that V-Sense is accurate in determining and differentiating various steering maneuvers, and is thus useful for a wide range of safety-assistance applications without additional sensors or infrastructure.
123 citations
"FullStop: Tracking unsafe stopping ..." refers background or methods in this paper
...We are in the process of exploring the use of other sensors such as the gyroscope to detect lane changes [13], in combination with our current set of features, to handle such scenarios of heavy optical occlusion....
[...]
...[13] uses a smartphone’s inertial sensors to detect unintentional lane departures of a moving vehicle....
[...]