Commit bfb9e11e authored by Diego García Fierro's avatar Diego García Fierro
Browse files

Update README.md

parent 180d4a35
......@@ -2,19 +2,19 @@
## Description ##
An Agent based Social Simulation System is proposed to analyse the way factors such as emotions, personality and stress interact and shape driving styles. Driver Agents are characterized by emotions, a personality profile, stress and distractions they might experience during the simulation, as well as several driving traits that define the way a driver behaves on the road: speed, acceleration, braking, steering and response time. A level system based on the intensity and the tendency of drivers towards each driving trait is defined in order to characterize driving styles.
An **Agent based Social Simulation System** is proposed to analyse the way factors such as **emotions**, **personality** and **stress** interact and shape **driving styles**. Driver Agents are characterized by emotions, a personality profile, stress and distractions they might experience during the simulation, as well as several driving traits that define the way a driver behaves on the road: speed, acceleration, braking, steering and response time. A level system based on the intensity and the tendency of drivers towards each driving trait is defined in order to characterize driving styles.
A model for accident probability estimation is included as well. The proposed model uses the Poisson Distribution to calculate accident probability. An accident dataset with accident reports of the city of Las Vegas (USA) between 2015 and 2018, used to obtain accident ocurrence rates for the Poisson Distribution application, is included in the repository.
A model for **accident probability estimation** is included as well. The proposed model uses the Poisson Distribution to calculate accident probability. An accident dataset with accident reports of the city of Las Vegas (USA) between 2015 and 2018, used to obtain accident ocurrence rates for the Poisson Distribution application, is included in the repository.
## Usage ##
Execution of the Simulation Model is carried out through the run.py file. Drivers taking part in the simulation can be modified in the general.ini configuration file, as well as steps for each simulation.
Execution of the Simulation Model is carried out through the *run.py* file. Drivers taking part in the simulation can be modified in the *general.ini* configuration file, as well as steps for each simulation.
Driver Agents are assigned random emotion, personality profile and stress values at the beginning of the simulation. Emotions, stress and distractions each Driver Agent is exposed to can change throughout the simulation. In each simulation step, the contribution that emotions and stress have to driving styles is considered. Personality is only evaluated at the beginning of the simulation.
Accident probability, calculated depending on the driving trait each Driver Agent stands out for (the one with more contributions and therefore the one that characterizes the Driver Agent's style the most) and the presence of distractions, is provided for each simulation step.
Data collected from each simulation is specified in both the add_collectors(self) function in the Model.py file and in the DriverCollector.py file.
Data collected from each simulation is specified in both the *add_collectors(self)* function in the *Model.py* file and in the *DriverCollector.py* file.
## Support ##
......
Supports Markdown
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment