- tomiedit
The estimation and analysis of road traffic represent the preliminary steps towards satisfying the current needs for smooth, safe, and green transportation. Therefore, effective traffic monitoring is an essential topic alongside the... more
The estimation and analysis of road traffic represent the preliminary steps towards satisfying the current needs for smooth, safe, and green transportation. Therefore, effective traffic monitoring is an essential topic alongside the planning of sustainable transportation systems and the development of new traffic management concepts. In contrast to classical traffic detection solutions, this study investigates the correlation between travelers' social activities and road traffic. The s's primary goal is to investigate the presence of the relationship between social activity and road traffic, which might allow an infrastructure-independent traffic monitoring technique as well. People's general activities at Point of Interest (POI) locations (measured as occupancy parameter) are correlated with traffic data so that, finally, proper proxys can be defined for link-level average traffic speed estimation. The method is tested and evaluated using real-world traffic and POI occupancy data from Budapest (District XI.). The results of the correlation investigation justify an indirect relationship between activity at POIs and road traffic, which holds promise for future practical applicability.
Research Interests:
Highly automated driving, let it be high level advanced driver assistance system (ADAS) or autonomous vehicle is an intensely developing area, in which trajectory planning is an important layer serving multiple needs of safety,... more
Highly automated driving, let it be high level advanced driver assistance system (ADAS) or autonomous vehicle is an intensely developing area, in which trajectory planning is an important layer serving multiple needs of safety, feasibility or passenger comfort. To provide dynamically feasible trajectory planning for vehicles based on different control goals one has to deal with the trade-off between numerical resources and correctness. The first part of the paper presents a solution which uses classical optimization based approach that generates the expected results though on the expense of computational effort, which is not acceptable in real-time environment. Two hybrid approaches are introduced in the second part, where the optimization is replaced or aided with artificial neural networks (ANN), trained with the data generated off-line by the first planner. Results show that the solution highly decreases the running time of the algorithm, with almost the same level of performance. The described algorithm can be used as a base for a generic dynamically feasible yet real-time solution.
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Several problems can be encountered in the design of autonomous vehicles. Their software is organized into three main layers: perception, planning, and actuation. The planning layer deals with the sort and long-term situation prediction,... more
Several problems can be encountered in the design of autonomous vehicles. Their software is organized into three main layers: perception, planning, and actuation. The planning layer deals with the sort and long-term situation prediction, which are crucial for intelligent vehicles. Whatever method is used to make forecasts, vehicles’ dynamic environment must be processed for accurate long-term forecasting. In the present article, a method is proposed to preprocess the dynamic environment in a freeway traffic situation. The method uses the structured data of surrounding vehicles and transforms it to an occupancy grid which a Convolutional Variational Autoencoder (CVAE) processes. The grids (2048 pixels) are compressed to a 64-dimensional latent vector by the encoder and reconstructed by the decoder. The output pixel intensities are interpreted as probabilities of the corresponding field is occupied by a vehicle. This method’s benefit is to preprocess the structured data of the dynamic...
Research Interests:
Research Interests:
Research Interests:
Research Interests:
The rapid growth of urbanization and the constant demand for mobility have put a great strain on transportation systems in cities. One of the major challenges in these areas is traffic congestion, particularly at signalized intersections.... more
The rapid growth of urbanization and the constant demand for mobility have put a great strain on transportation systems in cities. One of the major challenges in these areas is traffic congestion, particularly at signalized intersections. This problem not only leads to longer travel times for commuters, but also results in a significant increase in local and global emissions. The fixed cycle of traffic lights at these intersections is one of the primary reasons for this issue. To address these challenges, applying reinforcement learning to coordinating traffic light controllers has become a highly researched topic in the field of transportation engineering. This paper focuses on the traffic signal control problem, proposing a solution using a multi-agent deep Q-learning algorithm. This study introduces a novel rewarding concept in the multi-agent environment, as the reward schemes have yet to evolve in the following years with the advancement of techniques. The goal of this study is...
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Nonlinear optimization-based motion planning algorithms have been successfully used for dynamically feasible trajectory planning of road vehicles. However, the main drawback of these methods is their significant computational effort and... more
Nonlinear optimization-based motion planning algorithms have been successfully used for dynamically feasible trajectory planning of road vehicles. However, the main drawback of these methods is their significant computational effort and thus high runtime, which makes real-time application a complex problem. Addressing this field, this paper proposes an algorithm for fast simulation of road vehicle motion based on artificial neural networks that can be used in optimization-based trajectory planners. The neural networks are trained with supervised learning techniques to predict the future state of the vehicle based on its current state and driving inputs. Learning data is provided for a wide variety of randomly generated driving scenarios by simulation of a dynamic vehicle model. The realistic random driving maneuvers are created on the basis of piecewise linear travel velocity and road curvature profiles that are used for the planning of public roads. The trained neural networks are ...
Research Interests:
Planning the optimal trajectory of emergency avoidance maneuvers for highly automated vehicles is a complex task with many challenges. The algorithm needs to decrease accident risk by reducing the severity and keeping the car in a... more
Planning the optimal trajectory of emergency avoidance maneuvers for highly automated vehicles is a complex task with many challenges. The algorithm needs to decrease accident risk by reducing the severity and keeping the car in a controllable state. Optimal trajectory generation considering all aspects of vehicle and environment dynamics is numerically complex, especially if the object to be avoided is moving. This paper presents a hierarchical method for the avoidance of moving objects in an autonomous vehicle, where a reinforcement learning agent is responsible for local planning, while longitudinal and lateral control is performed by the low-level model-predictive controller and Stanley controllers. In the developed architecture, the agent is responsible for the optimization. It is trained in various scenarios to provide the necessary parameters for a polynomial-based path and a velocity profile in a neural network output. The vehicle performs only the first step of the trajecto...
Research Interests:
The paper presents real-world test cases of an optimal trajectory design solution that combines modern control techniques with machine learning. The first step of the current research is to train a reinforcement learning agent in a... more
The paper presents real-world test cases of an optimal trajectory design solution that combines modern control techniques with machine learning. The first step of the current research is to train a reinforcement learning agent in a simulated environment, where the conditions and the applied vehicle are modeled. System dynamics is described by a nonlinear single-track vehicle with dynamic wheel model. The designed trajectory is evaluated by driving the vehicle using a control loop. The reward of the method is based on the sum of different measures considering safety and passenger comfort. The proposed method forms a special one-step reinforcement learning task handled by Deep Deterministic Policy Gradient (DDPG) learning agent. As a result, the learning process provides a real-time neural-network-based motion planner and a tracking algorithm. The evaluation of the algorithm under real conditions is made by using an experimental test vehicle. The test setup contains a high precision GPS module, an automotive inertial sensor, an industrial PC, and communication interface devices. The test cases were performed on the ZalaZone automotive proving ground.
Research Interests:
Research Interests:
Research Interests:
Research Interests:
The real-time railway rescheduling problem is a crucial challenge for human operators since many factors have to be considered during decision making, from the positions and velocities of the vehicles to the different regulations of the... more
The real-time railway rescheduling problem is a crucial challenge for human operators since many factors have to be considered during decision making, from the positions and velocities of the vehicles to the different regulations of the individual railway companies. Thanks to that, human operators cannot be expected to provide optimal decisions in a particular situation. Based on the recent successes of multi-agent deep reinforcement learning in challenging control problems, it seems like a suitable choice for such a domain. Consequently, this paper proposes a multi-agent deep reinforcement learning-based approach with different state representational choices to solve the real-time railway rescheduling problem. Furthermore, comparing different methods, the proposed learning-based approaches outperform their competitions, such as the Monte Carlo tree search algorithm, which is utilized as a model-based planner, and also other learning-based methods that utilize different abstractions...
Research Interests:
The paper presents the modeling and control design of an electromechanical heavy-duty clutch actuator using gain-scheduled MPC and grid-based Linear Parameter Varying approaches. First, the nonlinear model of the electromechanical... more
The paper presents the modeling and control design of an electromechanical heavy-duty clutch actuator using gain-scheduled MPC and grid-based Linear Parameter Varying approaches. First, the nonlinear model of the electromechanical actuator is presented, then a third order quasi-Linear Parameter Varying representation of the system is derived, which takes the nonlinear characteristic of the diaphragm spring into account. Using the control-oriented model, a Linear Parameter Varying controller and a gain-scheduled Model Predictive Controller are designed, the latter of which serves as benchmark. The controllers have been implemented and tested in a model in the loop environment, where their performances have been compared concerning their rise-time, steady-state error, over-and undershoots, and robustness to the changes of the touch-point. The validation results show that the difference between the model predictive controllers is negligible in most cases, and they surpass the linear pa...
Research Interests:
Research Interests:
The paper presents multimodel state estimation with constrained filtering applicable in a road traffic situation. The state to be estimated describes the motion of a car observed from the ego-vehicle. Multiple motion modes of the target... more
The paper presents multimodel state estimation with constrained filtering applicable in a road traffic situation. The state to be estimated describes the motion of a car observed from the ego-vehicle. Multiple motion modes of the target vehicle are predefined and each is associated with a suitable constraining method. Both Kalman and particle filters are used to perform the estimations. The constrained filters are implemented in the Interacting Multiple Model structure. The outputs are the actual state of the observed vehicle and the motion mode that is in effect at every moment with an associated likelihood value. The performance of the proposed method is evaluated in a simulated environment.
Research Interests:
This work presents a powerful and intelligent driver agent, designed to operate in a preset highway situation using Policy Gradient Reinforcement Learning (RL) agent. Our goal is to create an agent that is capable of navigating safely in... more
This work presents a powerful and intelligent driver agent, designed to operate in a preset highway situation using Policy Gradient Reinforcement Learning (RL) agent. Our goal is to create an agent that is capable of navigating safely in changing highway traffic and successfully accomplish to get through the defined section keeping the reference speed. Meanwhile, creating a state representation that is capable of extracting information from images based on the actual highway situation. The algorithm uses Convolutional Neural Network (CNN) with Long-Short Term Memory (LSTM) layers as a function approximator for the agent with discrete action space on the control level, e.g., acceleration and lane change. Simulation of Urban MObility (SUMO), an open-source microscopic traffic simulator is chosen as our simulation environment. It is integrated with an open interface to interact with the agent in real-time. The agent can learn from numerous driving and highway situations that are created and fed to it. The representation becomes more general by randomizing and customizing the behavior of the other road users in the simulation, thus the experience of the agent can be much more diverse. The article briefly describes the modeling environment, the details on the learning agent, and the rewarding scheme. After evaluating the experiences gained from the training, some further plans and optimization ideas are briefed.
Research Interests:
Research Interests:
Research Interests:
Environment perception is one of the major challenges in the vehicle industry nowadays, as acknowledging the intentions of the surrounding traffic participants can profoundly decrease the occurrence of accidents. Consequently, this paper... more
Environment perception is one of the major challenges in the vehicle industry nowadays, as acknowledging the intentions of the surrounding traffic participants can profoundly decrease the occurrence of accidents. Consequently, this paper focuses on comparing different motion models, acknowledging their role in the performance of maneuver classification. In particular, this paper proposes utilizing the Interacting Multiple Model framework complemented with constrained Kalman filtering in this domain that enables the comparisons of the different motions models’ accuracy. The performance of the proposed method with different motion models is thoroughly evaluated in a simulation environment, including an observer and observed vehicle.
Research Interests:
This paper considers the object detection and tracking problem in a road traffic situation from a traffic participant’s perspective. The information source is an automotive radar which is attached to the ego vehicle. The scenario... more
This paper considers the object detection and tracking problem in a road traffic situation from a traffic participant’s perspective. The information source is an automotive radar which is attached to the ego vehicle. The scenario characteristics are varying object visibility due to occlusion and multiple detections of a vehicle during a scanning interval. The goal is to maintain and report the state of undetected though possibly present objects. The proposed algorithm is based on the multi-object Probability Hypothesis Density filter. Because the PHD filter has no memory, the estimate of the number of objects present can change abruptly due to erroneous detections. To reduce this effect, we model the occlusion of the object to calculate the state-dependent detection probability. Thus, the filter can maintain unnoticed but probably valid hypotheses for a more extended period. We use the sequential Monte Carlo method with clustering for implementing the filter. We distinguish between ...
Research Interests:
Research Interests:
Research Interests:
Research Interests:
This paper deals with the control design of an electromechanical shift actuator. First, to provide an environment where the controllers can be tuned and compared, a nonlinear model has been developed, then a linear state-space... more
This paper deals with the control design of an electromechanical shift actuator. First, to provide an environment where the controllers can be tuned and compared, a nonlinear model has been developed, then a linear state-space representation of the system has been derived. Besides tracking of a reference signal with high accuracy, the control design must consider the limitations of a typical, commercial actuator control unit, in particular real-time applicability. To achieve the controlling aims, a symmetric, optimum based, cascaded PID, a Linear Quadratic Regulator, a Linear Quadratic Gaussian controller, and a Model Predictive Controller have been developed. Three of the four developed methods can meet the given requirements. However, there are significant differences regarding the run-time of the algorithms.
Research Interests:
This paper presents a synergy of the Monte-Carlo tree search (MCTS) and a reinforcement learning (RL) based control strategy to achieve the position control of an electropneumatic gearbox actuator. Besides tracking the reference signal,... more
This paper presents a synergy of the Monte-Carlo tree search (MCTS) and a reinforcement learning (RL) based control strategy to achieve the position control of an electropneumatic gearbox actuator. Besides tracking the reference signal, there are qualitative requirements regarding the switching time and the overshoot, and there is also a necessity of reliable behavior in a wide range of operating conditions. By utilizing the domain-specific knowledge of a trained agent, the direction of the tree search can be controlled, hence the quality of the RL control can be further enhanced by the robustness of the MCTS algorithm.
Research Interests:
This paper deals with the control design of an electro-pneumatic gearbox actuator. The controller must be able to handle the highly nonlinear and unstable behavior of the system, while it also has to meet strict, partly contradictory... more
This paper deals with the control design of an electro-pneumatic gearbox actuator. The controller must be able to handle the highly nonlinear and unstable behavior of the system, while it also has to meet strict, partly contradictory requirements. The state-space representation of the actuator can be formulated as a quasi-Linear Parameter Varying system, thus a grid-based LPV/ℋ2 controller has been developed, which has been tested in a Model in the Loop environment. Based on the testing results, the controller proved to be a good trade-off between the requirements. Meanwhile, it has better overall performance than the widely used LTI control methods.
Research Interests:
Research Interests:
The paper deals with the modeling of an electropneumatic gearbox actuator. The objective of the research is to develop the nonlinear, mathematical model of an electropneumatic gearbox actuator, which can be used as a Model in The Loop... more
The paper deals with the modeling of an electropneumatic gearbox actuator. The objective of the research is to develop the nonlinear, mathematical model of an electropneumatic gearbox actuator, which can be used as a Model in The Loop environment for controller testing purposes, and to derivate a simplified, multi-state, linear model, which will be the basis of controller development.