Ziya Ercan

Conversion of a conventional electric automobile into an unmanned ground vehicle (UGV)

In this study, conversion procedure of a conventional electric automobile into an unmanned ground vehicle (UGV) is illustrated. This conversion process is divided into two main parts as, mechanical and electrical modifications. Interface circuit, interface software, additional power system, selection of the sensors and computer hardware are given in electrical modifications part. Similarly, design of braking and steering system, their computer simulations and strength analysis are given in mechanical modifications part. All these applications are illustrated on a conventional electric vehicle during this study.

Multi-sensor data fusion of DCM based orientation estimation for land vehicles

In this paper, an algorithm estimating orientation is implemented using Direction Cosine Matrix (DCM) method, chosen due to its linear process model and ease of use. Two Kalman filters were used to estimate the rotation matrix elements where the Euler Angles are easily computed. A rule based decision structure is used to choose the best measurement available in the system from GPS and digital compass. Also the dynamic motion of the vehicle is considered to overcome the slow response of the digital compass. The algorithm is tested with real time logged data set and a decision structure is developed to have the best Information provided from the multiple sensors. The algorithm is also tested under artificial GPS outages, performs successfully for both attitude and heading angles.

A new fuzzy speed control strategy considering lateral vehicle dynamics

This paper introduces a new speed control strategy for autonomous/semi-autonomous navigation of ground vehicles. Different from the previous studies, steering angle is considered in addition to speed error and integral of the speed error. Because of the highly nonlinear nature of vehicle model, fuzzy logic strategy is used for controller design. Vehicle modeling equations and used parameters are also illustrated in the paper. Simulations are carried out to verify and demonstrate the effectiveness of the new method over the classical one which does not consider the steering angle. Both of these two methods have similar performances when steering angle is relatively low. However, in a more aggressive steering scenario, classical approach fails and vehicle loses its yaw stability while our method still continues to track the speed with a stable yaw dynamics.

A machine learning approach for personalized autonomous lane change initiation and control

We study an algorithm that allows a vehicle to autonomously change lanes in a safe but personalized fashion without the drivers explicit initiation (e.g. activating the turn signals). Lane change initiation in autonomous driving is typically based on subjective rules, functions of the positions and relative velocities of surrounding vehicles. This approach is often arbitrary, and not easily adapted to the driving style preferences of an individual driver. Here we propose a data-driven modeling approach to capture the lane change decision behavior of human drivers. We collect data with a test vehicle in typical lane change situations and train classifiers to predict the instant of lane change initiation with respect to the preferences of a particular driver. We integrate this decision logic into a model predictive control (MPC) framework to create a more personalized autonomous lane change experience that satisfies safety and comfort constraints. We show the ability of the decision logic to reproduce and differentiate between two lane changing styles, and demonstrate the safety and effectiveness of the control framework through simulations.

Torque-based steering assistance for collision avoidance during lane changes

Lane changing is a cognitively demanding task that drivers perform frequently while driving on highways at high speed. This paper presents a predictive control framework which prevents collisions during lane change maneuvers. Approaches based on steering angle control are commonly used in the literature. These approaches may inadvertently take away the ultimate control authority of the driver. Also they involve a separate lower level steering control that requires subjective and complex steering feel calibration and steering torque constraints. The method proposed in this paper is based on a corrective torque formulation which incorporates the steering dynamics and driver response in the model. We tested the proposed controller in an experimental vehicle with a human driver. The results demonstrate that the proposed method avoids collision threats with minimum intervention.

A predictive control framework for torque-based steering assistance to improve safety in highway driving

ABSTRACT Haptic shared control framework opens up new perspectives on the design and implementation of the driver steering assistance systems which provide torque feedback to the driver in order to improve safety. While designing such a system, it is important to account for the human–machine interactions since the driver feels the feedback torque through the hand wheel. The controller should consider the drivers impact on the steering dynamics to achieve a better performance in terms of drivers acceptance and comfort. In this paper we present a predictive control framework which uses a model of driver-in-the-loop steering dynamics to optimise the torque intervention with respect to the drivers neuromuscular response. We first validate the system in simulations to compare the performance of the controller in nominal and model mismatch cases. Then we implement the controller in a test vehicle and perform experiments with a human driver. The results show the effectiveness of the proposed system in avoiding hazardous situations under different driver behaviours.

A new fuzzy speed planning method for safe navigation

This paper introduces a new speed planning strategy for autonomous navigation. Speed planning can be done considering lots of parameters using offline path planning strategies. However, in an obstacle avoidance scenario, which can be thought of as a dynamic path planning, avoidance strategy must work fast. That is why previous strategies generally concentrate only on steering maneuvers for obstacle avoidance. This paper concentrates on the speed planning part of the obstacle avoidance strategy. To this aim, a new fuzzy approach is developed for on-line speed planning with its purely reactive nature. Methodic simulations are carried out to verify and demonstrate the effectiveness of the new method over previous methods. The maneuver strategy for obstacle avoidance is the artificial potential field method.

Modeling, Identification, and Predictive Control of a Driver Steering Assistance System

This paper presents the design of a driver steering assistance system, which provides a corrective torque in order to guide the driver. While designing such a system, it is important to consider the interactions since the driver modifies the transfer function from the control input to the output of interest during a shared steering task. The novelty of our approach lies in the formulation of the predictive controller, which employs a model of the driver-in-the-loop steering dynamics, where an online parameter identification scheme is proposed to track the time-varying parameters of the process. An optimal guidance torque is calculated with respect to the level of interaction using the updated model. We validate the proposed approach by performing an experimental study with five participants in a guided lane, keeping task under different interaction behaviors of participants. The results show the system capability to adapt the control input based on the drivers acceptance on the torque intervention.

Unmanned Ground Vehicle Otonobil: Design, Perception, and Decision Algorithms

Unmanned ground vehicles (UGV) have been the subject of research in recent years due to their future prospective of solving the traffic congestion and improving the safety on roads while having a more energy-efficient profile. In this chapter, the first UGV of Turkey, Otonobil, will be introduced detailing especially on its hardware and software design architecture, the perception capabilities and decision algorithms used in obstacle avoidance, and autonomous goal-oriented docking. UGV Otonobil features a novel heuristic algorithm to avoid dynamic obstacles, and the vehicle is an open test-rig for studying several intelligent-vehicle technologies such as steer-by-wire, intelligent traction control, and further artificial intelligence algorithms for acting in real-traffic conditions.