Demo Abstract: Nericell - Using Mobile ... - Semantic Scholar
Demo Abstract: Nericell - Using Mobile Smartphones for Rich Monitoring of Road and Traffic Conditions Prashanth Mohan [email protected]
Ramachandran Ramjee Venkata N. Padmanabhan [email protected] [email protected]
Microsoft Research India, Bangalore
ABSTRACT We consider the problem of monitoring road and traffic conditions in a city. Prior work in this area has required the deployment of dedicated sensors on vehicles and/or on the roadside, or the tracking of mobile phones by service providers. Furthermore, prior work has largely focused on the developed world, with its relatively simple traffic flow patterns. In fact, traffic flow in cities of the developing regions, which comprise much of the world, tends to be much more complex owing to varied road conditions (e.g., potholed roads), chaotic traffic (e.g., a lot of braking and honking), and a heterogeneous mix of vehicles (2-wheelers, 3-wheelers, cars, buses, etc.). To monitor road and traffic conditions in such a setting, we present Nericell, a system that performs rich sensing by piggybacking on smartphones that users carry around with them. In this demo, we show the use of accelerometer to detect bumps and braking. We also use the microphone to enable honk detection. Nericell addresses several challenges including virtually reorienting the accelerometer on a phone that is at an arbitrary orientation, and performing honk detection and localization in an energy efficient manner.
Categories and Subject Descriptors C.m [Computer Systems Organization]: Miscellaneous— Mobile Sensing Systems
General Terms Algorithms, Design, Experimentation, Measurement
Keywords
Figure 1: Typical pothole on a roadway measuring approximately 1m in width developing world tend to be more varied because of various socio-economic reasons. Road quality tends to be variable, with bumpy roads and potholes being commonplace even in the heart of cities (see Figure 1). The flow of traffic can be chaotic, with little or no adherence to right of way protocols at some intersections and liberal use of honking. We present Nericell, a system for rich monitoring of road and traffic conditions that piggybacks on mobile smartphones that users carry when they drive a vehicle. A phone might include any or all of a microphone, camera, GPS, and accelerometer, each of which could be used for traffic sensing functions. In addition, the phone would include a cellular radio (e.g., GSM), possibly with data communication capabilities (e.g., GPRS or UMTS).
Sensing, Roads, Traffic, Mobile Phones
2. OVERVIEW OF NERICELL 1.
INTRODUCTION
Roads and vehicular traffic are a key part of the day-today lives of people. Therefore, monitoring their conditions has received a significant amount of attention. Prior work ([1, 2, 4, 6]) in this area has primarily focused on the developed world, where good roads and orderly traffic mean that the traffic conditions on a stretch of road can largely be characterized by the volume and speed of traffic flowing through it. In contrast, road and traffic conditions in the Copyright is held by the author/owner(s). SenSys’08, November 5–7, 2008, Raleigh, North Carolina, USA. ACM 978-1-59593-990-6/08/11.
Nericell utilises the sensors on mobile smartphones to detect braking, bumps and honking in the vicinity. We use the microphone to capture audio samples for honk detection. Additionally, we make use of a 3-axis accelerometer (which we assume to be part of the smartphone) to enable brake and bump detection. We define a canonical frame of reference, with the X axis pointing to the front of the vehicle, the Y axis to the side of the vehicle, and the Z axis vertically upwards. The measurement reported by the accelerometer is a function of the force exerted on its sensor mechanism. When the vehicle accelerates (which would represent a positive acceleration along X according to the textbook definition), our accelerometer would experience a
force pressing it backwards and hence report a negative acceleration along X. For the purposes of this description, we define aX as the acceleration along X axis, aY as the acceleration along Y axis, and aZ as the acceleration along Z axis. The accelerometer however may not be oriented in the same direction as the car. For this we “virtually reorient” the accelerometer to the direction of the car. More information on this can be obtained from our paper in SenSys 2008[5].
2.1 Braking Detection In general, braking would cause a surge in aX because the accelerometer would experience a force pushing it to the front. The surge can be significant even when the brake is applied at low speed. If a vehicle travelling at 10 kmph brakes to a halt in 1 second, that would result in an average of over 0.28g in aX and possibly much larger spikes. To detect the incidence of braking, we look for the mean of aX over a sliding window. We signal a braking event when there is a mean aX of 0.12g over a 4 second time window.
2.2 Bump Detection When a wheel enters a pothole, the wheel descends into the hole resulting in a sustained dip in the value of aZ until the wheel hits the bottom of the pothole, which causes a spike in the value of aZ . At high speeds, the surge in the value of aZ is very prominent. We found in our experiments that at low speeds, the surge in the value of aZ is not noticeable; however, the sustained dip in the values of aZ is evident. Thus, we utilize two bump detectors depending on the speed of the vehicle. At high speeds (> 25 kmph), we use the surge in aZ to detect bumps. This is identical to the z-peak heuristic proposed in [3]. When there is a spike along aZ greater than a threshold of 1.75g is classified as a suspect bump. At low speeds, we propose a new heuristic called zsus, which looks for a sustained dip in aZ (below a threshold of 0.8g) for at least 20 ms.
2.3 Honk Detection We implement a simple detector that takes 100ms audio samples, peforms a discrete Fourier transform on each such sample and looks for energy spikes. We define a spike as an instantaneous sample that is at least T times the mean, where T ranges between 5 and 10. As long as there are at least two spikes, including at least one spike in the 2.5 kHz to 4 kHz region corresponding to the region of highest human ear sensitivity, we classify the audio sample as including a honk.
3.
DEMO OVERVIEW
This is a demo of an associated paper to be presented at SenSys ’08 [5] which contains a more detailed description of the system. To demonstrate the Nericell system, we present the bump, brake and honk detection heuristics on an RC car. The demo setup consists of a Sparkfun WiTilt 3 axis accelerometer with a Bluetooth module, the HP iPAQ hw6965 smartphone and an RC car. We load the 3 axis accelerometer onto the RC car. The HP iPAQ smartphone receives the accelerometer’s readings over bluetooth. The detection algorithms run on the smartphone and detects events such as when the RC car brakes or runs over a bump (simulated with a cardboard strip). The microphone in the smartphone is used to gather audio samples and detect honking. The
Figure 2: Demo Setup thresholds provided in the previous section are pertinent to real world conditions. However, for our demo, the heuristics are scaled down to suit the size of an RC car.
4. REFERENCES [1] Intelligent Transportation Systems. http://www.its.dot.gov/. [2] OnStar by GM. http://www.onstar.com/. [3] J. Eriksson, L. Girod, B. Hull, R. Newton, S. Madden, and H. Balakrishnan. The Pothole Patrol: Using a Mobile Sensor Network for Road Surface Monitoring. In MobiSys, 2008. [4] B. Hull, V. Bychkovsky, K. Chen, M. Goraczko, A. Miu, E. Shih, Y. Zhang, H. Balakrishnan, and S. Madden. The CarTel Mobile Sensor Computing System. In SenSys, 2006. [5] P. Mohan, V. N. Padmanabhan, and R. Ramjee. Nericell: Rich Monitoring of Road and Traffic Conditions using Mobile Smartphones. In SenSys, 2008. To appear. [6] J. Yoon, B. Noble, and M. Liu. Surface Street Traffic Estimation. In MobiSys, 2007.
Ramachandran Ramjee Venkata N. Padmanabhan [email protected] [email protected]
Microsoft Research India, Bangalore
ABSTRACT We consider the problem of monitoring road and traffic conditions in a city. Prior work in this area has required the deployment of dedicated sensors on vehicles and/or on the roadside, or the tracking of mobile phones by service providers. Furthermore, prior work has largely focused on the developed world, with its relatively simple traffic flow patterns. In fact, traffic flow in cities of the developing regions, which comprise much of the world, tends to be much more complex owing to varied road conditions (e.g., potholed roads), chaotic traffic (e.g., a lot of braking and honking), and a heterogeneous mix of vehicles (2-wheelers, 3-wheelers, cars, buses, etc.). To monitor road and traffic conditions in such a setting, we present Nericell, a system that performs rich sensing by piggybacking on smartphones that users carry around with them. In this demo, we show the use of accelerometer to detect bumps and braking. We also use the microphone to enable honk detection. Nericell addresses several challenges including virtually reorienting the accelerometer on a phone that is at an arbitrary orientation, and performing honk detection and localization in an energy efficient manner.
Categories and Subject Descriptors C.m [Computer Systems Organization]: Miscellaneous— Mobile Sensing Systems
General Terms Algorithms, Design, Experimentation, Measurement
Keywords
Figure 1: Typical pothole on a roadway measuring approximately 1m in width developing world tend to be more varied because of various socio-economic reasons. Road quality tends to be variable, with bumpy roads and potholes being commonplace even in the heart of cities (see Figure 1). The flow of traffic can be chaotic, with little or no adherence to right of way protocols at some intersections and liberal use of honking. We present Nericell, a system for rich monitoring of road and traffic conditions that piggybacks on mobile smartphones that users carry when they drive a vehicle. A phone might include any or all of a microphone, camera, GPS, and accelerometer, each of which could be used for traffic sensing functions. In addition, the phone would include a cellular radio (e.g., GSM), possibly with data communication capabilities (e.g., GPRS or UMTS).
Sensing, Roads, Traffic, Mobile Phones
2. OVERVIEW OF NERICELL 1.
INTRODUCTION
Roads and vehicular traffic are a key part of the day-today lives of people. Therefore, monitoring their conditions has received a significant amount of attention. Prior work ([1, 2, 4, 6]) in this area has primarily focused on the developed world, where good roads and orderly traffic mean that the traffic conditions on a stretch of road can largely be characterized by the volume and speed of traffic flowing through it. In contrast, road and traffic conditions in the Copyright is held by the author/owner(s). SenSys’08, November 5–7, 2008, Raleigh, North Carolina, USA. ACM 978-1-59593-990-6/08/11.
Nericell utilises the sensors on mobile smartphones to detect braking, bumps and honking in the vicinity. We use the microphone to capture audio samples for honk detection. Additionally, we make use of a 3-axis accelerometer (which we assume to be part of the smartphone) to enable brake and bump detection. We define a canonical frame of reference, with the X axis pointing to the front of the vehicle, the Y axis to the side of the vehicle, and the Z axis vertically upwards. The measurement reported by the accelerometer is a function of the force exerted on its sensor mechanism. When the vehicle accelerates (which would represent a positive acceleration along X according to the textbook definition), our accelerometer would experience a
force pressing it backwards and hence report a negative acceleration along X. For the purposes of this description, we define aX as the acceleration along X axis, aY as the acceleration along Y axis, and aZ as the acceleration along Z axis. The accelerometer however may not be oriented in the same direction as the car. For this we “virtually reorient” the accelerometer to the direction of the car. More information on this can be obtained from our paper in SenSys 2008[5].
2.1 Braking Detection In general, braking would cause a surge in aX because the accelerometer would experience a force pushing it to the front. The surge can be significant even when the brake is applied at low speed. If a vehicle travelling at 10 kmph brakes to a halt in 1 second, that would result in an average of over 0.28g in aX and possibly much larger spikes. To detect the incidence of braking, we look for the mean of aX over a sliding window. We signal a braking event when there is a mean aX of 0.12g over a 4 second time window.
2.2 Bump Detection When a wheel enters a pothole, the wheel descends into the hole resulting in a sustained dip in the value of aZ until the wheel hits the bottom of the pothole, which causes a spike in the value of aZ . At high speeds, the surge in the value of aZ is very prominent. We found in our experiments that at low speeds, the surge in the value of aZ is not noticeable; however, the sustained dip in the values of aZ is evident. Thus, we utilize two bump detectors depending on the speed of the vehicle. At high speeds (> 25 kmph), we use the surge in aZ to detect bumps. This is identical to the z-peak heuristic proposed in [3]. When there is a spike along aZ greater than a threshold of 1.75g is classified as a suspect bump. At low speeds, we propose a new heuristic called zsus, which looks for a sustained dip in aZ (below a threshold of 0.8g) for at least 20 ms.
2.3 Honk Detection We implement a simple detector that takes 100ms audio samples, peforms a discrete Fourier transform on each such sample and looks for energy spikes. We define a spike as an instantaneous sample that is at least T times the mean, where T ranges between 5 and 10. As long as there are at least two spikes, including at least one spike in the 2.5 kHz to 4 kHz region corresponding to the region of highest human ear sensitivity, we classify the audio sample as including a honk.
3.
DEMO OVERVIEW
This is a demo of an associated paper to be presented at SenSys ’08 [5] which contains a more detailed description of the system. To demonstrate the Nericell system, we present the bump, brake and honk detection heuristics on an RC car. The demo setup consists of a Sparkfun WiTilt 3 axis accelerometer with a Bluetooth module, the HP iPAQ hw6965 smartphone and an RC car. We load the 3 axis accelerometer onto the RC car. The HP iPAQ smartphone receives the accelerometer’s readings over bluetooth. The detection algorithms run on the smartphone and detects events such as when the RC car brakes or runs over a bump (simulated with a cardboard strip). The microphone in the smartphone is used to gather audio samples and detect honking. The
Figure 2: Demo Setup thresholds provided in the previous section are pertinent to real world conditions. However, for our demo, the heuristics are scaled down to suit the size of an RC car.
4. REFERENCES [1] Intelligent Transportation Systems. http://www.its.dot.gov/. [2] OnStar by GM. http://www.onstar.com/. [3] J. Eriksson, L. Girod, B. Hull, R. Newton, S. Madden, and H. Balakrishnan. The Pothole Patrol: Using a Mobile Sensor Network for Road Surface Monitoring. In MobiSys, 2008. [4] B. Hull, V. Bychkovsky, K. Chen, M. Goraczko, A. Miu, E. Shih, Y. Zhang, H. Balakrishnan, and S. Madden. The CarTel Mobile Sensor Computing System. In SenSys, 2006. [5] P. Mohan, V. N. Padmanabhan, and R. Ramjee. Nericell: Rich Monitoring of Road and Traffic Conditions using Mobile Smartphones. In SenSys, 2008. To appear. [6] J. Yoon, B. Noble, and M. Liu. Surface Street Traffic Estimation. In MobiSys, 2007.
