SlideShare una empresa de Scribd logo
1 de 26
Descargar para leer sin conexión
BAYESIAN RISK ASSESSMENT OF
AUTONOMOUS VEHICLES
Christos Katrakazas
Mohammed Quddus
Wen-Hua Chen*
Transport Studies Group
School of Civil and Building Engineering
*Department of Aeronautics and Automobile Engineering
Loughborough University
NORTHMOST 01: ITS-Leeds Monday 12th Dec.
Overview
 Introduction to the problem
 Bayesian & Dynamic Bayesian Networks (DBN)
 DBN models and risk assessment of autonomous vehicles
- Variables, estimation of probabilities and inference
 Preliminary findings
 Potential contribution
3
Introduction
Human error is responsible for causing 75 – 90% traffic accidents
Examples:
• Blind-spots & line of sight
• Risk perception
• Reaction time
• Impaired driving
• Fails to look properly
• Excessive/inappropriate speed
Removing the human element from
the task of driving
Potential Solution?
Autonomous vehicles
Road to Autonomy
Potential obstacles?
- Reliability
- High quality data
- Perception horizon
How could Transport
Professional help?
4
© European Commission
Roadmap for automated driving
5
Robotics
 Expensive sensors
 Real-time effectiveness
 Lack of context
Collision Prediction (vehicle-level)
In-vehicle sensors
Dangerous road user
6
Transport Engineering
 Aggregated data
 Location-based variables
 Spatio-temporal risk
Could network-level collision predication in transport engineering be
integrated to vehicle-level risk assessment of autonomous vehicles?
- Bayesian Inference?
Collision Prediction (network-level)
Dangerous road segment
Classification
Real-time traffic data
Bayesian Networks
 Directed Acyclic Probabilistic
Graphs
 Every node represents a random
variable
 Edges represent probabilistic
dependencies or influences
 Joint Probability Distribution
shows how a situation is
modelled (e.g. the probabilistic
relationship between the
variables of the whole system)
7
Bayesian Networks
• Suitable for learning causal
relationships
• Ideal representation for combining
prior knowledge and data
• Help in modelling noisy systems
• Can handle situations where data is
incomplete
BUT
Are applied for events in a particular
point in time!
8
Dynamic Bayesian Networks (DBN)
 Bayesian Networks used to model a
system that dynamically changes or
evolves over time
 Probabilistic reasoning over time
 How do the variables affect each
other over time?
 Requirements for DBNs:
1. A prior probability P(x1)
2. A state-transition function P(xt|xt-1)
3. An observation function P(Yt|xt)
Time slice
9
Dynamic Bayesian Networks (DBN)
1. A prior (initial) probability
distribution P(x1) in the beginning of
the process;
2. A state-transition function P(xt|xt-1)
specifies time dependencies between
states/variables;
3. An observation function P(Yt|xt)
Specifies dependencies of
observation nodes regarding to
other nodes at time slice t.
10
Time slice
Dynamic Bayesian Network (DBN): Example
Raint-1 P(Raint-1)
True (T) 0.7
False (F) 0.3
Raint P(Umbrellat|Raint)
T 0.9
F 0.1
Rain : Hidden Variable
Umbrella : Observed Variable
11
Research Question
How could fundamental principles of robotics and transport
engineering be integrated in addressing research challenges
associated with real-time crash prediction of autonomous vehicles?
 Act proactively for the ego-vehicle
 Improve real-time prediction by using network-level hint
 Take traffic environment into account
 May reduce the need for expensive (“super”- accurate) sensor measurements
Potential improvements?
Modelling crash prediction in real-time
Required variables:
 Network-level Risk (CRN): “Is the road segment on which the vehicle
travels dangerous or not?”
 Vehicle-level Risk (CRV): “Are the vehicles in the vicinity of the ego-
vehicle dangerous or not?”
 Vehicle Kinematics (K): “How likely is that the vehicles will follow the
same course according to a physical model of motion?”
 Sensor Measurements (Z): “How likely is that the measurements from
the sensors are giving the correct values?”
How are the variables connected?
Observations
(Z)
Kinematics
(K)
Crash Risk
Vehicle-Level
(CRV)
Crash Risk
Network-Level
(CRN)
What happens on the road segment
influences the behaviour of the vehicles
If a situation between
vehicles is dangerous,
their motion will be
affected
The motion of the vehicles is depicted in the
sensors’ observations
Variable relationship depicted as a DBN
t t + 1 t+2
Figure: Dynamic Bayesian
Network
Markov State Space model
Multi-vehicle dependencies
Single vehicle dependencies
 Use traffic flow parameters to estimate the risk of an accident
happening in real-time
 Compare & Contrast traffic conditions just before an accident with
normal conditions
Data: Highways England & DfT
• 15-min Traffic flow data (HATRIS JTDB)
• Historical Accident data (STATS 19)
• Traffic microsimulation (PTV VISSIM) -> 30second traffic data
Method : Machine learning classifiers (i.e. SVMs, RVMs, Random
Forests, k-Nearest Neighbours)
Network – Level Risk
 Represents the probability of a crash happening between two
vehicles
 Needs a well-calibrated metric or risk indicator
Data
 Sensor measurements, Maps, Vehicle trajectories
 Methods
 Unscented Kalman Filter for sensor data fusion, Time-to-
collision metrics
 Problems: Efficient data fusion, crashes in real-world
environments
Vehicle – Level Risk
Safe and dangerous vehicle contexts
Which of the vehicle trajectories end
up in a collision?
Vehicle – Level Risk
𝑓𝐾 = 𝑓(TTCn
t−1
)
= ቊ
1: dangerous 𝑖𝑓 TTCn
t
< 𝐶𝑟𝑖𝑡𝑖𝑐𝑎𝑙 𝑇𝑇𝐶
0: 𝑠𝑎𝑓𝑒; 𝑜𝑡ℎ𝑒𝑟𝑤𝑖𝑠𝑒
Kinematics/ Vehicle motion
Kinematics
• Kinematics variable describes the probability that the vehicle will
follow a certain course according to the context.
• Uses information on position, heading and speed to distinguish
between contexts
Kinematics/ Vehicle motion
Kinematics
Bicycle model
Compromise between bicycle model estimations
and context thresholds
Accuracy of the sensors’ system
Sensor measurements
• Each measurement from the sensors contains only partial
information about the environment
• This variable (Z) describes the probability that the sensor
readings correspond correctly to the real values of the
attributes that are measured
Sensor Measurements
Correct measurements probability
Sensor Measurements
𝑃 Τ𝑍 𝑛
𝑡
𝐾 𝑛
𝑡
~ 𝑆𝑡𝑢𝑑𝑒𝑛𝑡 𝐶 𝑇
𝐾 𝑛
𝑡
, 𝜎2
𝛪, 𝜈
where C is a rectangular matrix that selects entries from
the kinematic (physical state), ν are the degrees of
freedom, Ι is the identity matrix and σ is related to the
accuracy of the sensor system.
Inference
t t + 1 t+2
𝑷 𝑪𝑹𝑽 𝒏
𝒕
= 𝒅 𝑪𝑹𝑽 𝑵
𝒕−𝟏
𝑲 𝑵
𝒕−𝟏
𝑪𝑹𝑵 𝒏
𝒕
> λ
Preliminary Findings:
Vehicle-level risk estimation
𝑷 𝑪𝑹𝑽 𝒏
𝒕
= 𝒅 𝑪𝑹𝑽 𝑵
𝒕−𝟏
𝑲 𝑵
𝒕−𝟏
𝑪𝑹𝑵 𝒏
𝒕
and assuming 6 vehicles are sensed
by the ego-vehicle
With network-level hint
σ 𝒏=𝟏
𝑵
(𝒇 𝑲 𝒏
= 𝟏) + σ 𝒏=𝟏
𝑵
(𝒇 𝑪𝑹𝑽 𝒏
= 𝟏) + σ 𝒏=𝟏
𝑵
(𝒇 𝑪𝑹𝑵 𝒏
= 𝟏)
𝑵
=
𝟏+𝟏+𝟏
𝟔
= 𝟎. 𝟓
Without network-level hint
σ 𝒏=𝟏
𝑵
(𝒇 𝑲 𝒏
= 𝟏) + σ 𝒏=𝟏
𝑵
(𝒇 𝑪𝑹𝑽 𝒏
= 𝟏)
𝑵
=
𝟏 + 𝟏
𝟔
= 𝟎. 𝟑𝟑
By simply adding a function checking the network-level collision
risk, hazardous vehicle identification is potentially improved!
25
Potential contribution
 Improve real-time effectiveness of
vehicle-level collision prediction by
making use of network-level risk
- Knowing the road segment
where an accident is likely to
happen
- Find faster which car is going
to trigger the accident in this
road segment
 Make AVs drive in a human-like cautious
way in road segments which are flagged
dangerous (e.g reduce speed)
 Assist obstructed or low-cost AV sensor’
systems.
Inspiring Winners Since 1909
Thank you!
Christos Katrakazas
c.katrakazas@lboro.ac.uk

Más contenido relacionado

La actualidad más candente

Computer Vision for Beginners
Computer Vision for BeginnersComputer Vision for Beginners
Computer Vision for Beginners
Sanghamitra Deb
 

La actualidad más candente (20)

Using Bayesian Optimization to Tune Machine Learning Models
Using Bayesian Optimization to Tune Machine Learning ModelsUsing Bayesian Optimization to Tune Machine Learning Models
Using Bayesian Optimization to Tune Machine Learning Models
 
Sensor fusion of LiDAR and Camera for real time object detection - talk version
Sensor fusion of LiDAR and Camera for real time object detection - talk versionSensor fusion of LiDAR and Camera for real time object detection - talk version
Sensor fusion of LiDAR and Camera for real time object detection - talk version
 
Discrete event-simulation
Discrete event-simulationDiscrete event-simulation
Discrete event-simulation
 
Cv_Chap 4 Segmentation
Cv_Chap 4 SegmentationCv_Chap 4 Segmentation
Cv_Chap 4 Segmentation
 
Semi-Supervised Learning
Semi-Supervised LearningSemi-Supervised Learning
Semi-Supervised Learning
 
Interpretability of Convolutional Neural Networks - Eva Mohedano - UPC Barcel...
Interpretability of Convolutional Neural Networks - Eva Mohedano - UPC Barcel...Interpretability of Convolutional Neural Networks - Eva Mohedano - UPC Barcel...
Interpretability of Convolutional Neural Networks - Eva Mohedano - UPC Barcel...
 
capsule network
capsule networkcapsule network
capsule network
 
Machine Learning - Simple Linear Regression
Machine Learning - Simple Linear RegressionMachine Learning - Simple Linear Regression
Machine Learning - Simple Linear Regression
 
PCA and SVD in brief
PCA and SVD in briefPCA and SVD in brief
PCA and SVD in brief
 
Object tracking a survey
Object tracking a surveyObject tracking a survey
Object tracking a survey
 
Visual Object Tracking: review
Visual Object Tracking: reviewVisual Object Tracking: review
Visual Object Tracking: review
 
A Gentle Introduction to the EM Algorithm
A Gentle Introduction to the EM AlgorithmA Gentle Introduction to the EM Algorithm
A Gentle Introduction to the EM Algorithm
 
Iterative Closest Point Algorithm - analysis and implementation
Iterative Closest Point Algorithm - analysis and implementationIterative Closest Point Algorithm - analysis and implementation
Iterative Closest Point Algorithm - analysis and implementation
 
Using synthetic data for computer vision model training
Using synthetic data for computer vision model trainingUsing synthetic data for computer vision model training
Using synthetic data for computer vision model training
 
Deep Reinforcement Learning: Q-Learning
Deep Reinforcement Learning: Q-LearningDeep Reinforcement Learning: Q-Learning
Deep Reinforcement Learning: Q-Learning
 
MIT 6.S091: Introduction to Deep Reinforcement Learning (Deep RL) by Lex Fridman
MIT 6.S091: Introduction to Deep Reinforcement Learning (Deep RL) by Lex FridmanMIT 6.S091: Introduction to Deep Reinforcement Learning (Deep RL) by Lex Fridman
MIT 6.S091: Introduction to Deep Reinforcement Learning (Deep RL) by Lex Fridman
 
Computer Vision for Beginners
Computer Vision for BeginnersComputer Vision for Beginners
Computer Vision for Beginners
 
Eigenface For Face Recognition
Eigenface For Face RecognitionEigenface For Face Recognition
Eigenface For Face Recognition
 
Multiple object detection
Multiple object detectionMultiple object detection
Multiple object detection
 
Time series forecasting with machine learning
Time series forecasting with machine learningTime series forecasting with machine learning
Time series forecasting with machine learning
 

Destacado

Wenyuan xu Minrui yan can you trust autonomous vehicles_slides_liu_final
Wenyuan xu Minrui yan can you trust autonomous vehicles_slides_liu_finalWenyuan xu Minrui yan can you trust autonomous vehicles_slides_liu_final
Wenyuan xu Minrui yan can you trust autonomous vehicles_slides_liu_final
PacSecJP
 
Designing Roads for AVs (autonomous vehicles)
Designing Roads for AVs (autonomous vehicles)Designing Roads for AVs (autonomous vehicles)
Designing Roads for AVs (autonomous vehicles)
Jeffrey Funk
 

Destacado (20)

Real time traffic management - challenges and solutions
Real time traffic management - challenges and solutionsReal time traffic management - challenges and solutions
Real time traffic management - challenges and solutions
 
Proportionally fair scheduling for traffic light networks
Proportionally fair scheduling for traffic light networksProportionally fair scheduling for traffic light networks
Proportionally fair scheduling for traffic light networks
 
A new theory of lane selection on highways
A new theory of lane selection on highwaysA new theory of lane selection on highways
A new theory of lane selection on highways
 
Agent based car following model for heterogeneities of platoon driving with v...
Agent based car following model for heterogeneities of platoon driving with v...Agent based car following model for heterogeneities of platoon driving with v...
Agent based car following model for heterogeneities of platoon driving with v...
 
Capacity maximising traffic signal control policies
Capacity maximising traffic signal control policiesCapacity maximising traffic signal control policies
Capacity maximising traffic signal control policies
 
Understanding quality of life operationalization in policy - lessons from uk ...
Understanding quality of life operationalization in policy - lessons from uk ...Understanding quality of life operationalization in policy - lessons from uk ...
Understanding quality of life operationalization in policy - lessons from uk ...
 
A bus ride with Foucalt
A bus ride with FoucaltA bus ride with Foucalt
A bus ride with Foucalt
 
Critical issues in estimating human exposure to traffic related air pollution...
Critical issues in estimating human exposure to traffic related air pollution...Critical issues in estimating human exposure to traffic related air pollution...
Critical issues in estimating human exposure to traffic related air pollution...
 
An Ontology-Based Intelligent Speed Adaptation System for Autonomous Cars
An Ontology-Based Intelligent Speed Adaptation System for Autonomous CarsAn Ontology-Based Intelligent Speed Adaptation System for Autonomous Cars
An Ontology-Based Intelligent Speed Adaptation System for Autonomous Cars
 
Creative industries flexibility of work and travel
Creative industries flexibility of work and travelCreative industries flexibility of work and travel
Creative industries flexibility of work and travel
 
2014 MATC Spring Lecture Series: Chris Schwarz
2014 MATC Spring Lecture Series: Chris Schwarz2014 MATC Spring Lecture Series: Chris Schwarz
2014 MATC Spring Lecture Series: Chris Schwarz
 
Autonomous Vehicles
Autonomous VehiclesAutonomous Vehicles
Autonomous Vehicles
 
Study HERE SBD - How autonomous vehicles could relieve or worsen traffic cong...
Study HERE SBD - How autonomous vehicles could relieve or worsen traffic cong...Study HERE SBD - How autonomous vehicles could relieve or worsen traffic cong...
Study HERE SBD - How autonomous vehicles could relieve or worsen traffic cong...
 
2016- A Year in Review of the Development of Autonomous vehicles
2016- A Year in Review of the Development of Autonomous vehicles2016- A Year in Review of the Development of Autonomous vehicles
2016- A Year in Review of the Development of Autonomous vehicles
 
Autonomous Vehicles and Reducing GHG
Autonomous Vehicles and Reducing GHGAutonomous Vehicles and Reducing GHG
Autonomous Vehicles and Reducing GHG
 
Millennials on the move - life transitions and Australian millennials
Millennials on the move - life transitions and Australian millennialsMillennials on the move - life transitions and Australian millennials
Millennials on the move - life transitions and Australian millennials
 
Connected & Autonomous vehicles: cybersecurity on a grand scale v1
Connected & Autonomous vehicles: cybersecurity on a grand scale v1Connected & Autonomous vehicles: cybersecurity on a grand scale v1
Connected & Autonomous vehicles: cybersecurity on a grand scale v1
 
Autonomous Vehicles: the Intersection of Robotics and Artificial Intelligence
Autonomous Vehicles: the Intersection of Robotics and Artificial IntelligenceAutonomous Vehicles: the Intersection of Robotics and Artificial Intelligence
Autonomous Vehicles: the Intersection of Robotics and Artificial Intelligence
 
Wenyuan xu Minrui yan can you trust autonomous vehicles_slides_liu_final
Wenyuan xu Minrui yan can you trust autonomous vehicles_slides_liu_finalWenyuan xu Minrui yan can you trust autonomous vehicles_slides_liu_final
Wenyuan xu Minrui yan can you trust autonomous vehicles_slides_liu_final
 
Designing Roads for AVs (autonomous vehicles)
Designing Roads for AVs (autonomous vehicles)Designing Roads for AVs (autonomous vehicles)
Designing Roads for AVs (autonomous vehicles)
 

Similar a Bayesian risk assessment of autonomous vehicles

Machine Learning and Optimization For Traffic and Emergency ...
Machine Learning and Optimization For Traffic and Emergency ...Machine Learning and Optimization For Traffic and Emergency ...
Machine Learning and Optimization For Traffic and Emergency ...
butest
 
Deep graph convolutional networks for incident driven traffic speed prediction
Deep graph convolutional networks for incident driven traffic speed predictionDeep graph convolutional networks for incident driven traffic speed prediction
Deep graph convolutional networks for incident driven traffic speed prediction
ivaderivader
 

Similar a Bayesian risk assessment of autonomous vehicles (20)

Fowe Thesis Full
Fowe Thesis FullFowe Thesis Full
Fowe Thesis Full
 
Study of statistical models for route prediction algorithms in vanet
Study of statistical models for route prediction algorithms in vanetStudy of statistical models for route prediction algorithms in vanet
Study of statistical models for route prediction algorithms in vanet
 
Machine Learning and Optimization For Traffic and Emergency ...
Machine Learning and Optimization For Traffic and Emergency ...Machine Learning and Optimization For Traffic and Emergency ...
Machine Learning and Optimization For Traffic and Emergency ...
 
Deep graph convolutional networks for incident driven traffic speed prediction
Deep graph convolutional networks for incident driven traffic speed predictionDeep graph convolutional networks for incident driven traffic speed prediction
Deep graph convolutional networks for incident driven traffic speed prediction
 
Prediction of traveller information and route choice
Prediction of traveller information and route choicePrediction of traveller information and route choice
Prediction of traveller information and route choice
 
Automated Traffic System Control
Automated Traffic System ControlAutomated Traffic System Control
Automated Traffic System Control
 
The Joy of SLAM
The Joy of SLAMThe Joy of SLAM
The Joy of SLAM
 
5100termproj422
5100termproj4225100termproj422
5100termproj422
 
CREATING DATA OUTPUTS FROM MULTI AGENT TRAFFIC MICRO SIMULATION TO ASSIMILATI...
CREATING DATA OUTPUTS FROM MULTI AGENT TRAFFIC MICRO SIMULATION TO ASSIMILATI...CREATING DATA OUTPUTS FROM MULTI AGENT TRAFFIC MICRO SIMULATION TO ASSIMILATI...
CREATING DATA OUTPUTS FROM MULTI AGENT TRAFFIC MICRO SIMULATION TO ASSIMILATI...
 
Where Next
Where NextWhere Next
Where Next
 
Pedestrian behavior/intention modeling for autonomous driving V
Pedestrian behavior/intention modeling for autonomous driving VPedestrian behavior/intention modeling for autonomous driving V
Pedestrian behavior/intention modeling for autonomous driving V
 
CREATING DATA OUTPUTS FROM MULTI AGENT TRAFFIC MICRO SIMULATION TO ASSIMILATI...
CREATING DATA OUTPUTS FROM MULTI AGENT TRAFFIC MICRO SIMULATION TO ASSIMILATI...CREATING DATA OUTPUTS FROM MULTI AGENT TRAFFIC MICRO SIMULATION TO ASSIMILATI...
CREATING DATA OUTPUTS FROM MULTI AGENT TRAFFIC MICRO SIMULATION TO ASSIMILATI...
 
Using Genetic Algorithm for Shortest Path Selection with Real Time Traffic Flow
Using Genetic Algorithm for Shortest Path Selection with Real Time Traffic FlowUsing Genetic Algorithm for Shortest Path Selection with Real Time Traffic Flow
Using Genetic Algorithm for Shortest Path Selection with Real Time Traffic Flow
 
Multi-agent approach to resource allocation inautonomous vehicle fleet
Multi-agent approach to resource allocation inautonomous vehicle fleetMulti-agent approach to resource allocation inautonomous vehicle fleet
Multi-agent approach to resource allocation inautonomous vehicle fleet
 
A Generic Agent Model Towards Comparing Resource Allocation Approaches to On-...
A Generic Agent Model Towards Comparing Resource Allocation Approaches to On-...A Generic Agent Model Towards Comparing Resource Allocation Approaches to On-...
A Generic Agent Model Towards Comparing Resource Allocation Approaches to On-...
 
Wherecamp Navigation Conference 2015 - The unintelligent swarm
Wherecamp Navigation Conference 2015 - The unintelligent swarmWherecamp Navigation Conference 2015 - The unintelligent swarm
Wherecamp Navigation Conference 2015 - The unintelligent swarm
 
TRAFFIC FORECAST FOR INTELLECTUAL TRANSPORTATION SYSTEM USING MACHINE LEARNING
TRAFFIC FORECAST FOR INTELLECTUAL TRANSPORTATION SYSTEM USING MACHINE LEARNINGTRAFFIC FORECAST FOR INTELLECTUAL TRANSPORTATION SYSTEM USING MACHINE LEARNING
TRAFFIC FORECAST FOR INTELLECTUAL TRANSPORTATION SYSTEM USING MACHINE LEARNING
 
IRJET - A Review on Pedestrian Behavior Prediction for Intelligent Transport ...
IRJET - A Review on Pedestrian Behavior Prediction for Intelligent Transport ...IRJET - A Review on Pedestrian Behavior Prediction for Intelligent Transport ...
IRJET - A Review on Pedestrian Behavior Prediction for Intelligent Transport ...
 
Predictive Data Dissemination in Vanet
Predictive Data Dissemination in VanetPredictive Data Dissemination in Vanet
Predictive Data Dissemination in Vanet
 
Smart Traffic Congestion Control System: Leveraging Machine Learning for Urba...
Smart Traffic Congestion Control System: Leveraging Machine Learning for Urba...Smart Traffic Congestion Control System: Leveraging Machine Learning for Urba...
Smart Traffic Congestion Control System: Leveraging Machine Learning for Urba...
 

Más de Institute for Transport Studies (ITS)

Social networks, activities, and travel - building links to understand behaviour
Social networks, activities, and travel - building links to understand behaviourSocial networks, activities, and travel - building links to understand behaviour
Social networks, activities, and travel - building links to understand behaviour
Institute for Transport Studies (ITS)
 
Rail freight in Japan - track access
Rail freight in Japan - track accessRail freight in Japan - track access
Rail freight in Japan - track access
Institute for Transport Studies (ITS)
 
Shipping by the crowd - empirical analysis of operations and behavior
Shipping by the crowd - empirical analysis of operations and behaviorShipping by the crowd - empirical analysis of operations and behavior
Shipping by the crowd - empirical analysis of operations and behavior
Institute for Transport Studies (ITS)
 
Towards an effective transitioning of public transport system in Ghana
Towards an effective transitioning of public transport system in GhanaTowards an effective transitioning of public transport system in Ghana
Towards an effective transitioning of public transport system in Ghana
Institute for Transport Studies (ITS)
 

Más de Institute for Transport Studies (ITS) (20)

Transport Projects Aimed at Fostering Economic Growth – experience in the UK ...
Transport Projects Aimed at Fostering Economic Growth – experience in the UK ...Transport Projects Aimed at Fostering Economic Growth – experience in the UK ...
Transport Projects Aimed at Fostering Economic Growth – experience in the UK ...
 
BA Geography with Transport Studies at the University of Leeds
BA Geography with Transport Studies at the University of LeedsBA Geography with Transport Studies at the University of Leeds
BA Geography with Transport Studies at the University of Leeds
 
Highways Benchmarking - Accelerating Impact
Highways Benchmarking - Accelerating ImpactHighways Benchmarking - Accelerating Impact
Highways Benchmarking - Accelerating Impact
 
Using telematics data to research traffic related air pollution
Using telematics data to research traffic related air pollutionUsing telematics data to research traffic related air pollution
Using telematics data to research traffic related air pollution
 
Masters Dissertation Posters 2017
Masters Dissertation Posters 2017Masters Dissertation Posters 2017
Masters Dissertation Posters 2017
 
Institute for Transport Studies - Masters Open Day 2017
Institute for Transport Studies - Masters Open Day 2017Institute for Transport Studies - Masters Open Day 2017
Institute for Transport Studies - Masters Open Day 2017
 
London's Crossrail Scheme - its evolution, governance, financing and challenges
London's Crossrail Scheme  - its evolution, governance, financing and challengesLondon's Crossrail Scheme  - its evolution, governance, financing and challenges
London's Crossrail Scheme - its evolution, governance, financing and challenges
 
Secretary of State Visit
Secretary of State VisitSecretary of State Visit
Secretary of State Visit
 
Business model innovation for electrical vehicle futures
Business model innovation for electrical vehicle futuresBusiness model innovation for electrical vehicle futures
Business model innovation for electrical vehicle futures
 
A clustering method based on repeated trip behaviour to identify road user cl...
A clustering method based on repeated trip behaviour to identify road user cl...A clustering method based on repeated trip behaviour to identify road user cl...
A clustering method based on repeated trip behaviour to identify road user cl...
 
Cars cars everywhere
Cars cars everywhereCars cars everywhere
Cars cars everywhere
 
Annual Review 2015-16 - University of leeds
Annual Review 2015-16 - University of leedsAnnual Review 2015-16 - University of leeds
Annual Review 2015-16 - University of leeds
 
Social networks, activities, and travel - building links to understand behaviour
Social networks, activities, and travel - building links to understand behaviourSocial networks, activities, and travel - building links to understand behaviour
Social networks, activities, and travel - building links to understand behaviour
 
Rail freight in Japan - track access
Rail freight in Japan - track accessRail freight in Japan - track access
Rail freight in Japan - track access
 
Shipping by the crowd - empirical analysis of operations and behavior
Shipping by the crowd - empirical analysis of operations and behaviorShipping by the crowd - empirical analysis of operations and behavior
Shipping by the crowd - empirical analysis of operations and behavior
 
Towards an effective transitioning of public transport system in Ghana
Towards an effective transitioning of public transport system in GhanaTowards an effective transitioning of public transport system in Ghana
Towards an effective transitioning of public transport system in Ghana
 
City regions, transport and devolution
City regions, transport and devolutionCity regions, transport and devolution
City regions, transport and devolution
 
Mobility in deprived settlements - walking and the built environment
Mobility in deprived settlements - walking and the built environmentMobility in deprived settlements - walking and the built environment
Mobility in deprived settlements - walking and the built environment
 
Incorporating social influence into hybrid choice models
Incorporating social influence into hybrid choice modelsIncorporating social influence into hybrid choice models
Incorporating social influence into hybrid choice models
 
Managing the introduction of electric vehicles into the car fleet of Scotland
Managing the introduction of electric vehicles into the car fleet of ScotlandManaging the introduction of electric vehicles into the car fleet of Scotland
Managing the introduction of electric vehicles into the car fleet of Scotland
 

Último

Call Girls Bangalore Just Call 👗 7737669865 👗 Top Class Call Girl Service Ban...
Call Girls Bangalore Just Call 👗 7737669865 👗 Top Class Call Girl Service Ban...Call Girls Bangalore Just Call 👗 7737669865 👗 Top Class Call Girl Service Ban...
Call Girls Bangalore Just Call 👗 7737669865 👗 Top Class Call Girl Service Ban...
amitlee9823
 
一比一原版(PU学位证书)普渡大学毕业证学历认证加急办理
一比一原版(PU学位证书)普渡大学毕业证学历认证加急办理一比一原版(PU学位证书)普渡大学毕业证学历认证加急办理
一比一原版(PU学位证书)普渡大学毕业证学历认证加急办理
ezgenuh
 
Call Girls Kadugodi Just Call 👗 7737669865 👗 Top Class Call Girl Service Bang...
Call Girls Kadugodi Just Call 👗 7737669865 👗 Top Class Call Girl Service Bang...Call Girls Kadugodi Just Call 👗 7737669865 👗 Top Class Call Girl Service Bang...
Call Girls Kadugodi Just Call 👗 7737669865 👗 Top Class Call Girl Service Bang...
amitlee9823
 
Sanjay Nagar Call Girls: 🍓 7737669865 🍓 High Profile Model Escorts | Bangalor...
Sanjay Nagar Call Girls: 🍓 7737669865 🍓 High Profile Model Escorts | Bangalor...Sanjay Nagar Call Girls: 🍓 7737669865 🍓 High Profile Model Escorts | Bangalor...
Sanjay Nagar Call Girls: 🍓 7737669865 🍓 High Profile Model Escorts | Bangalor...
amitlee9823
 
Call Girls in Patel Nagar, Delhi 💯 Call Us 🔝9953056974 🔝 Escort Service
Call Girls in Patel Nagar, Delhi 💯 Call Us 🔝9953056974 🔝 Escort ServiceCall Girls in Patel Nagar, Delhi 💯 Call Us 🔝9953056974 🔝 Escort Service
Call Girls in Patel Nagar, Delhi 💯 Call Us 🔝9953056974 🔝 Escort Service
9953056974 Low Rate Call Girls In Saket, Delhi NCR
 
+971565801893>>SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN ABUDHABI,DUBAI MA...
+971565801893>>SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN ABUDHABI,DUBAI MA...+971565801893>>SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN ABUDHABI,DUBAI MA...
+971565801893>>SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN ABUDHABI,DUBAI MA...
Health
 
一比一原版(UdeM学位证书)蒙特利尔大学毕业证学历认证怎样办
一比一原版(UdeM学位证书)蒙特利尔大学毕业证学历认证怎样办一比一原版(UdeM学位证书)蒙特利尔大学毕业证学历认证怎样办
一比一原版(UdeM学位证书)蒙特利尔大学毕业证学历认证怎样办
ezgenuh
 
Escorts Service Rajajinagar ☎ 7737669865☎ Book Your One night Stand (Bangalore)
Escorts Service Rajajinagar ☎ 7737669865☎ Book Your One night Stand (Bangalore)Escorts Service Rajajinagar ☎ 7737669865☎ Book Your One night Stand (Bangalore)
Escorts Service Rajajinagar ☎ 7737669865☎ Book Your One night Stand (Bangalore)
amitlee9823
 

Último (20)

Call Girls Bangalore Just Call 👗 7737669865 👗 Top Class Call Girl Service Ban...
Call Girls Bangalore Just Call 👗 7737669865 👗 Top Class Call Girl Service Ban...Call Girls Bangalore Just Call 👗 7737669865 👗 Top Class Call Girl Service Ban...
Call Girls Bangalore Just Call 👗 7737669865 👗 Top Class Call Girl Service Ban...
 
一比一原版(PU学位证书)普渡大学毕业证学历认证加急办理
一比一原版(PU学位证书)普渡大学毕业证学历认证加急办理一比一原版(PU学位证书)普渡大学毕业证学历认证加急办理
一比一原版(PU学位证书)普渡大学毕业证学历认证加急办理
 
Is Your BMW PDC Malfunctioning Discover How to Easily Reset It
Is Your BMW PDC Malfunctioning Discover How to Easily Reset ItIs Your BMW PDC Malfunctioning Discover How to Easily Reset It
Is Your BMW PDC Malfunctioning Discover How to Easily Reset It
 
Call Girls Kadugodi Just Call 👗 7737669865 👗 Top Class Call Girl Service Bang...
Call Girls Kadugodi Just Call 👗 7737669865 👗 Top Class Call Girl Service Bang...Call Girls Kadugodi Just Call 👗 7737669865 👗 Top Class Call Girl Service Bang...
Call Girls Kadugodi Just Call 👗 7737669865 👗 Top Class Call Girl Service Bang...
 
Sanjay Nagar Call Girls: 🍓 7737669865 🍓 High Profile Model Escorts | Bangalor...
Sanjay Nagar Call Girls: 🍓 7737669865 🍓 High Profile Model Escorts | Bangalor...Sanjay Nagar Call Girls: 🍓 7737669865 🍓 High Profile Model Escorts | Bangalor...
Sanjay Nagar Call Girls: 🍓 7737669865 🍓 High Profile Model Escorts | Bangalor...
 
How To Troubleshoot Mercedes Blind Spot Assist Inoperative Error
How To Troubleshoot Mercedes Blind Spot Assist Inoperative ErrorHow To Troubleshoot Mercedes Blind Spot Assist Inoperative Error
How To Troubleshoot Mercedes Blind Spot Assist Inoperative Error
 
What Does The Engine Malfunction Reduced Power Message Mean For Your BMW X5
What Does The Engine Malfunction Reduced Power Message Mean For Your BMW X5What Does The Engine Malfunction Reduced Power Message Mean For Your BMW X5
What Does The Engine Malfunction Reduced Power Message Mean For Your BMW X5
 
John Deere 7430 7530 Tractors Diagnostic Service Manual W.pdf
John Deere 7430 7530 Tractors Diagnostic Service Manual W.pdfJohn Deere 7430 7530 Tractors Diagnostic Service Manual W.pdf
John Deere 7430 7530 Tractors Diagnostic Service Manual W.pdf
 
John Deere 335 375 385 435 Service Repair Manual
John Deere 335 375 385 435 Service Repair ManualJohn Deere 335 375 385 435 Service Repair Manual
John Deere 335 375 385 435 Service Repair Manual
 
(ISHITA) Call Girls Service Jammu Call Now 8617697112 Jammu Escorts 24x7
(ISHITA) Call Girls Service Jammu Call Now 8617697112 Jammu Escorts 24x7(ISHITA) Call Girls Service Jammu Call Now 8617697112 Jammu Escorts 24x7
(ISHITA) Call Girls Service Jammu Call Now 8617697112 Jammu Escorts 24x7
 
Call Girls in Patel Nagar, Delhi 💯 Call Us 🔝9953056974 🔝 Escort Service
Call Girls in Patel Nagar, Delhi 💯 Call Us 🔝9953056974 🔝 Escort ServiceCall Girls in Patel Nagar, Delhi 💯 Call Us 🔝9953056974 🔝 Escort Service
Call Girls in Patel Nagar, Delhi 💯 Call Us 🔝9953056974 🔝 Escort Service
 
John deere 425 445 455 Maitenance Manual
John deere 425 445 455 Maitenance ManualJohn deere 425 445 455 Maitenance Manual
John deere 425 445 455 Maitenance Manual
 
+971565801893>>SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN ABUDHABI,DUBAI MA...
+971565801893>>SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN ABUDHABI,DUBAI MA...+971565801893>>SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN ABUDHABI,DUBAI MA...
+971565801893>>SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN ABUDHABI,DUBAI MA...
 
一比一原版(UdeM学位证书)蒙特利尔大学毕业证学历认证怎样办
一比一原版(UdeM学位证书)蒙特利尔大学毕业证学历认证怎样办一比一原版(UdeM学位证书)蒙特利尔大学毕业证学历认证怎样办
一比一原版(UdeM学位证书)蒙特利尔大学毕业证学历认证怎样办
 
How To Fix Mercedes Benz Anti-Theft Protection Activation Issue
How To Fix Mercedes Benz Anti-Theft Protection Activation IssueHow To Fix Mercedes Benz Anti-Theft Protection Activation Issue
How To Fix Mercedes Benz Anti-Theft Protection Activation Issue
 
Call Girls in Malviya Nagar Delhi 💯 Call Us 🔝9205541914 🔝( Delhi) Escorts Ser...
Call Girls in Malviya Nagar Delhi 💯 Call Us 🔝9205541914 🔝( Delhi) Escorts Ser...Call Girls in Malviya Nagar Delhi 💯 Call Us 🔝9205541914 🔝( Delhi) Escorts Ser...
Call Girls in Malviya Nagar Delhi 💯 Call Us 🔝9205541914 🔝( Delhi) Escorts Ser...
 
What Causes BMW Chassis Stabilization Malfunction Warning To Appear
What Causes BMW Chassis Stabilization Malfunction Warning To AppearWhat Causes BMW Chassis Stabilization Malfunction Warning To Appear
What Causes BMW Chassis Stabilization Malfunction Warning To Appear
 
Escorts Service Rajajinagar ☎ 7737669865☎ Book Your One night Stand (Bangalore)
Escorts Service Rajajinagar ☎ 7737669865☎ Book Your One night Stand (Bangalore)Escorts Service Rajajinagar ☎ 7737669865☎ Book Your One night Stand (Bangalore)
Escorts Service Rajajinagar ☎ 7737669865☎ Book Your One night Stand (Bangalore)
 
Why Does My Porsche Cayenne's Exhaust Sound So Loud
Why Does My Porsche Cayenne's Exhaust Sound So LoudWhy Does My Porsche Cayenne's Exhaust Sound So Loud
Why Does My Porsche Cayenne's Exhaust Sound So Loud
 
(INDIRA) Call Girl Surat Call Now 8250077686 Surat Escorts 24x7
(INDIRA) Call Girl Surat Call Now 8250077686 Surat Escorts 24x7(INDIRA) Call Girl Surat Call Now 8250077686 Surat Escorts 24x7
(INDIRA) Call Girl Surat Call Now 8250077686 Surat Escorts 24x7
 

Bayesian risk assessment of autonomous vehicles

  • 1. BAYESIAN RISK ASSESSMENT OF AUTONOMOUS VEHICLES Christos Katrakazas Mohammed Quddus Wen-Hua Chen* Transport Studies Group School of Civil and Building Engineering *Department of Aeronautics and Automobile Engineering Loughborough University NORTHMOST 01: ITS-Leeds Monday 12th Dec.
  • 2. Overview  Introduction to the problem  Bayesian & Dynamic Bayesian Networks (DBN)  DBN models and risk assessment of autonomous vehicles - Variables, estimation of probabilities and inference  Preliminary findings  Potential contribution
  • 3. 3 Introduction Human error is responsible for causing 75 – 90% traffic accidents Examples: • Blind-spots & line of sight • Risk perception • Reaction time • Impaired driving • Fails to look properly • Excessive/inappropriate speed Removing the human element from the task of driving Potential Solution? Autonomous vehicles
  • 4. Road to Autonomy Potential obstacles? - Reliability - High quality data - Perception horizon How could Transport Professional help? 4 © European Commission Roadmap for automated driving
  • 5. 5 Robotics  Expensive sensors  Real-time effectiveness  Lack of context Collision Prediction (vehicle-level) In-vehicle sensors Dangerous road user
  • 6. 6 Transport Engineering  Aggregated data  Location-based variables  Spatio-temporal risk Could network-level collision predication in transport engineering be integrated to vehicle-level risk assessment of autonomous vehicles? - Bayesian Inference? Collision Prediction (network-level) Dangerous road segment Classification Real-time traffic data
  • 7. Bayesian Networks  Directed Acyclic Probabilistic Graphs  Every node represents a random variable  Edges represent probabilistic dependencies or influences  Joint Probability Distribution shows how a situation is modelled (e.g. the probabilistic relationship between the variables of the whole system) 7
  • 8. Bayesian Networks • Suitable for learning causal relationships • Ideal representation for combining prior knowledge and data • Help in modelling noisy systems • Can handle situations where data is incomplete BUT Are applied for events in a particular point in time! 8
  • 9. Dynamic Bayesian Networks (DBN)  Bayesian Networks used to model a system that dynamically changes or evolves over time  Probabilistic reasoning over time  How do the variables affect each other over time?  Requirements for DBNs: 1. A prior probability P(x1) 2. A state-transition function P(xt|xt-1) 3. An observation function P(Yt|xt) Time slice 9
  • 10. Dynamic Bayesian Networks (DBN) 1. A prior (initial) probability distribution P(x1) in the beginning of the process; 2. A state-transition function P(xt|xt-1) specifies time dependencies between states/variables; 3. An observation function P(Yt|xt) Specifies dependencies of observation nodes regarding to other nodes at time slice t. 10 Time slice
  • 11. Dynamic Bayesian Network (DBN): Example Raint-1 P(Raint-1) True (T) 0.7 False (F) 0.3 Raint P(Umbrellat|Raint) T 0.9 F 0.1 Rain : Hidden Variable Umbrella : Observed Variable 11
  • 12. Research Question How could fundamental principles of robotics and transport engineering be integrated in addressing research challenges associated with real-time crash prediction of autonomous vehicles?  Act proactively for the ego-vehicle  Improve real-time prediction by using network-level hint  Take traffic environment into account  May reduce the need for expensive (“super”- accurate) sensor measurements Potential improvements?
  • 13. Modelling crash prediction in real-time Required variables:  Network-level Risk (CRN): “Is the road segment on which the vehicle travels dangerous or not?”  Vehicle-level Risk (CRV): “Are the vehicles in the vicinity of the ego- vehicle dangerous or not?”  Vehicle Kinematics (K): “How likely is that the vehicles will follow the same course according to a physical model of motion?”  Sensor Measurements (Z): “How likely is that the measurements from the sensors are giving the correct values?”
  • 14. How are the variables connected? Observations (Z) Kinematics (K) Crash Risk Vehicle-Level (CRV) Crash Risk Network-Level (CRN) What happens on the road segment influences the behaviour of the vehicles If a situation between vehicles is dangerous, their motion will be affected The motion of the vehicles is depicted in the sensors’ observations
  • 15. Variable relationship depicted as a DBN t t + 1 t+2 Figure: Dynamic Bayesian Network Markov State Space model Multi-vehicle dependencies Single vehicle dependencies
  • 16.  Use traffic flow parameters to estimate the risk of an accident happening in real-time  Compare & Contrast traffic conditions just before an accident with normal conditions Data: Highways England & DfT • 15-min Traffic flow data (HATRIS JTDB) • Historical Accident data (STATS 19) • Traffic microsimulation (PTV VISSIM) -> 30second traffic data Method : Machine learning classifiers (i.e. SVMs, RVMs, Random Forests, k-Nearest Neighbours) Network – Level Risk
  • 17.  Represents the probability of a crash happening between two vehicles  Needs a well-calibrated metric or risk indicator Data  Sensor measurements, Maps, Vehicle trajectories  Methods  Unscented Kalman Filter for sensor data fusion, Time-to- collision metrics  Problems: Efficient data fusion, crashes in real-world environments Vehicle – Level Risk
  • 18. Safe and dangerous vehicle contexts Which of the vehicle trajectories end up in a collision? Vehicle – Level Risk 𝑓𝐾 = 𝑓(TTCn t−1 ) = ቊ 1: dangerous 𝑖𝑓 TTCn t < 𝐶𝑟𝑖𝑡𝑖𝑐𝑎𝑙 𝑇𝑇𝐶 0: 𝑠𝑎𝑓𝑒; 𝑜𝑡ℎ𝑒𝑟𝑤𝑖𝑠𝑒
  • 19. Kinematics/ Vehicle motion Kinematics • Kinematics variable describes the probability that the vehicle will follow a certain course according to the context. • Uses information on position, heading and speed to distinguish between contexts
  • 20. Kinematics/ Vehicle motion Kinematics Bicycle model Compromise between bicycle model estimations and context thresholds Accuracy of the sensors’ system
  • 21. Sensor measurements • Each measurement from the sensors contains only partial information about the environment • This variable (Z) describes the probability that the sensor readings correspond correctly to the real values of the attributes that are measured Sensor Measurements
  • 22. Correct measurements probability Sensor Measurements 𝑃 Τ𝑍 𝑛 𝑡 𝐾 𝑛 𝑡 ~ 𝑆𝑡𝑢𝑑𝑒𝑛𝑡 𝐶 𝑇 𝐾 𝑛 𝑡 , 𝜎2 𝛪, 𝜈 where C is a rectangular matrix that selects entries from the kinematic (physical state), ν are the degrees of freedom, Ι is the identity matrix and σ is related to the accuracy of the sensor system.
  • 23. Inference t t + 1 t+2 𝑷 𝑪𝑹𝑽 𝒏 𝒕 = 𝒅 𝑪𝑹𝑽 𝑵 𝒕−𝟏 𝑲 𝑵 𝒕−𝟏 𝑪𝑹𝑵 𝒏 𝒕 > λ
  • 24. Preliminary Findings: Vehicle-level risk estimation 𝑷 𝑪𝑹𝑽 𝒏 𝒕 = 𝒅 𝑪𝑹𝑽 𝑵 𝒕−𝟏 𝑲 𝑵 𝒕−𝟏 𝑪𝑹𝑵 𝒏 𝒕 and assuming 6 vehicles are sensed by the ego-vehicle With network-level hint σ 𝒏=𝟏 𝑵 (𝒇 𝑲 𝒏 = 𝟏) + σ 𝒏=𝟏 𝑵 (𝒇 𝑪𝑹𝑽 𝒏 = 𝟏) + σ 𝒏=𝟏 𝑵 (𝒇 𝑪𝑹𝑵 𝒏 = 𝟏) 𝑵 = 𝟏+𝟏+𝟏 𝟔 = 𝟎. 𝟓 Without network-level hint σ 𝒏=𝟏 𝑵 (𝒇 𝑲 𝒏 = 𝟏) + σ 𝒏=𝟏 𝑵 (𝒇 𝑪𝑹𝑽 𝒏 = 𝟏) 𝑵 = 𝟏 + 𝟏 𝟔 = 𝟎. 𝟑𝟑 By simply adding a function checking the network-level collision risk, hazardous vehicle identification is potentially improved!
  • 25. 25 Potential contribution  Improve real-time effectiveness of vehicle-level collision prediction by making use of network-level risk - Knowing the road segment where an accident is likely to happen - Find faster which car is going to trigger the accident in this road segment  Make AVs drive in a human-like cautious way in road segments which are flagged dangerous (e.g reduce speed)  Assist obstructed or low-cost AV sensor’ systems.
  • 26. Inspiring Winners Since 1909 Thank you! Christos Katrakazas c.katrakazas@lboro.ac.uk