Abstract

KAAPLAX is the result of a student research project at Chalmers University of Technology, Gothenburg. 

The goal of this project work is to develop a solution for an assistance system for an autonomous vehicle. The unique and novel challenge of this active safety system is to predict bicyclist’s behavior by using existing sensors and technology. Implementing an algorithm in a virtual environment which prevents crashes in a critical defined situation is the main objective of this project. 

The critical defined situation is a non-signalized T-intersection with a bicycle lane crossing the path of the side road. Participants in the chosen scenario are the autonomous vehicle, which is fully equipped with newest sensors and the developed threat assessment algorithm. A bicycle driving on the bicycle lane will enter the intersection from the same direction as the autonomous vehicle. A non-autonomous car coming towards with a given trajectory from naturalistic data studies will enter the intersection from the opposite direction. 

The autonomous vehicle will turn left into the intersection with a real trajectory from naturalistic data studies. Before the turning maneuver the autonomous car will switch on the indicators. The bicycle will recognize the indicators and will start braking after a literature based reaction time. Based on experimental datasets from traditional and E-bicycles, the autonomous car takes action to avoid a crash with the bicycle or the oncoming traffic. A decision making algorithm is developed to show possible actions of an autonomous vehicle in a real world intersection to avoid a crash on different stopping points. 

The main objective for the project was to ensure that the autonomous emergency brakes (AEB) on autonomous cars will work in order to prevent the crash between a bicyclist and a car. These kinds of projects are necessary with the focus rapidly shifting towards building autonomous cars. The results of the KAAPLAX algorithm have been very positive and promising. The threat assessment and decision making algorithm worked flawlessly in every scenario as was defined by the team for the project. A total of 65 simulations each were carried out for each of the scenario listed with codes written for the algorithm and the results then visualized in a virtual environment. The AEB was actuated in every case where the autonomous vehicle assessed that there was a potential crash. 

Keywords: Autonomous vehicle, T-intersection, traditional and e-bicycles, threat assessment and decision making algorithm, Autonomous Emergency Brake. 

Simulation results

1:Simulation results based on the developed threat assessment algorithm to predict cyclists behaviour in a given intersection from Klaus_Gabel on Vimeo.

2: Simulation results based on the developed threat assessment algorithm to predict cyclists behaviour in a given intersection from Klaus_Gabel on Vimeo.

3: Simulation results based on the developed threat assessment algorithm to predict cyclists behaviour in a given intersection from Klaus_Gabel on Vimeo.

KAAPLAX presentation slides

Introduction

Man has come a long way from the invention of wheel thousands of years ago. The car today has become an indispensable source of travel for covering short and long distances. However, man now aims to reach the next summit in terms of personal travel. Autonomous or self driving cars are slowly turning out to be the future path in terms of automotive technologies. With the current advancements in this field from manufacturing companies like Volvo, BMW, Mercedes etc. and even Google, the autonomous cars looks all set to debut within the next five years. 
This is supposed to make travel safer and at the same time would allow the people to carry on with their work without having to worry about the traffic and the stress associated with driving. Several technologies like Adaptive Cruise Control, Lane Keeping Assist, Blind Spot Detection, Automatic Emergency Braking etc. and many others have been developed and implemented in the cars running on the roads today. These technologies are the transition between manual to autonomous driving. In other words, these are like the first steps towards fully autonomous cars. 
The one of the main USP for autonomous cars are that it will make driving safer as the reaction time of the car with its algorithms and sensors is faster than a human being. Autonomous cars are not just supposed to protect the people inside the car but also the people outside which include other car drivers, pedestrians and cyclists. 
This project aims at improving the safety for the cyclists by identifying the behavior of the cyclists, the critical scenarios and decision making algorithms which will make the car brake in the event that an imminent crash is sensed by the system. We have used sensor fusion methods which include using a LIDAR for continuous monitoring of the cyclist once within the view of the car and the autonomous braking depending on the inputs received by the model. 
Autonomous cars can become a reality only if people have confidence that it will keep all the road users safe. This project is an effort in that direction which would ultimately help in the acceptance of the autonomous cars. We hope to simulate critical scenarios and show that the car would brake every time there is a chance to collide with the cyclist. Our approach to solve and deliver on this project is what will be explored in the next topics.

What is KAAPLAX?

KAAPLAX is the result of a student research project at Chalmers University of Technology, Göteborg. The goal of the project is to develop a solution for an assistance system for an autonomous vehicles. 

The unique and novel challenge of this active safety system is to predict bicyclist’s behavior by using existing sensors and technology. Implementing an algorithm in a virtual environment which prevents crashes in a critical defined situation, within the end of December 2015.

From left side: Klaus, Abhay, Abhishek, Prateek, Lin, Akshay, Xuming

K A A P L A X