| 1. |
- Chandru, Rajashekar, 1991, et al.
(author)
-
Safe autonomous lane changes in dense traffic
- 2017
-
In: IEEE International Conference on Intelligent Transportation Systems-ITSC. - 2153-0009. - 9781538615263 ; 2018-March
-
Conference paper (peer-reviewed)abstract
- Lane change manoeuvres are complex driving manoeuvres to automate since the vehicle has to anticipate and adapt to intentions of several surrounding vehicles. Selecting a suitable gap to move/merge into the adjacent lane and performing the lane change can be challenging, especially in dense traffic. Existing gap selection methods tend to be either cautious or opportunistic, both of which directly affect the overall availability and safety of the autonomous feature. In this paper we present a method which enables the autonomous vehicles to increase the availability of lane change manoeuvres by reducing the required margins to ensure a safe manoeuvre. The required safety margins are first calculated by making use of the steering and braking capability of the vehicle. It is then shown that this method can be used to perform autonomous lane changes in dense traffic situations with small inter-vehicle gaps. The proposed solution is evaluated by using Model Predictive Control (MPC) to plan and execute the complete motion trajectory.
|
|
| 2. |
- Kovaceva, Jordanka, 1980, et al.
(author)
-
Identification of aggressive driving from naturalistic data in car-following situations
- 2020
-
In: Journal of Safety Research. - : Elsevier BV. - 0022-4375. ; 73, s. 225-234
-
Journal article (peer-reviewed)abstract
- Introduction: Aggressive driving has been associated as one of the causes for crashes, sometimes with very serious consequences. The objective of this study is to investigate the possibility of identifying aggressive driving in car-following situations on motorways by simple jerk metrics derived from naturalistic data. Method: We investigate two jerk metrics, one for large positive jerk and the other for large negative jerk, when drivers are operating the gas and brake pedal, respectively. Results: The results obtained from naturalistic data from five countries in Europe show that the drivers from different countries have a significantly different number of large positive and large negative jerks. Male drivers operate the vehicle with significantly larger number of negative jerks compared to female drivers. The validation of the jerk metrics in identifying aggressive driving is performed by tailgating (following a leading vehicle in a close proximity) and by a violator/non-violator categorization derived from self-reported questionnaires. Our study shows that the identification of aggressive driving could be reinforced by the number of large negative jerks, given that the drivers are tailgating, or by the number of large positive jerks, given that the drivers are categorized as violators. Practical applications: The possibility of understanding, classifying, and quantifying aggressive driving behavior and driving styles with higher risk for accidents can be used for the development of driver support and coaching programs that promote driver safety and are enabled by the vast collection of driving data from modern in-vehicle monitoring and smartphone technology.
|
|
| 3. |
- Aubeck, Franz, et al.
(author)
-
Performance Comparison of Real-Time Solver Implementations for Powertrain Nonlinear Energy Management Optimization with MPC
- 2020
-
In: European Control Conference 2020, ECC 2020. ; , s. 483-490
-
Conference paper (peer-reviewed)abstract
- During recent years, commercial vehicle manufacturers have introduced legally required and advanced predictive functionalities into their vehicles. By applying modern supervisory Model Predictive Controllers (MPCs) for vehicle energy management and by also using the live traffic data, it is possible to reduce efficiently the fuel consumption and pollutant emissions. The general objective is to minimize fuel consumption by optimizing the speed and gear without affecting the overall travel time. In this work, the objective is to compare the performance of two modern optimization solvers, the interior-point solver FORCES Pro and ACADO that uses the active-set solver qpOASES for solving the energy management problem without controlling the powertrain states like gear selection. The novel energy management cost function consists of 6 optimization variables, different distance based discretization levels with velocity and time as state variables, whereby a solution is obtained within a few milliseconds at each iteration. Such controllers are applied to a 40-ton truck simulator and verified with several driving cycles. The novel approach achieves a fuel consumption decrease by more than 7% and less than 2% increase in execution time with the FORCES Pro solver compared to non-predictive rule based strategies. Also, the FORCES Pro solver was found to be 50 times faster than ACADO. The innovative software and solver framework provide an efficient energy management solution for future conventional and hybrid drivetrains.
|
|
| 4. |
- Börve, Erik, 1998, et al.
(author)
-
Interaction-Aware Trajectory Prediction and Planning in Dense Highway Traffic using Distributed Model Predictive Control
- 2023
-
In: Proceedings of the IEEE Conference on Decision and Control. - 2576-2370 .- 0743-1546. ; , s. 6124-6129
-
Conference paper (peer-reviewed)abstract
- In this paper we treat optimal trajectory planning for an autonomous vehicle (AV) operating in dense traffic, where vehicles closely interact with each other. To tackle this problem, we present a novel framework that couples trajectory prediction and planning in multi-agent environments, using distributed model predictive control. A demonstration of our framework is presented in simulation, employing a trajectory planner using non-linear model predictive control. We analyze performance and convergence of our framework, subject to different prediction errors. The results indicate that the obtained locally optimal solutions are improved, compared with decoupled prediction and planning.
|
|
| 5. |
- Chen, Mo, et al.
(author)
-
Energy-Efficient and Safe-Separation Operation for Successive Trains
- 2023
-
In: IEEE Conference on Intelligent Transportation Systems, Proceedings, ITSC. - 2153-0017 .- 2153-0009. ; , s. 845-851
-
Conference paper (peer-reviewed)abstract
- Energy-efficient and safe-separation during train operation is of great significance for urban rail transit systems, particularly on lines with high traffic density. This paper integrates the two issues and proposes a general cooperative operation method for successive trains with no limit on the number of trains. The cooperation problem is formulated as an optimal control problem and then solved as a nonlinear program. By simultaneously optimizing the speed profiles of each train, the total traction energy of the multi-train system can be minimized, ensuring safety by imposing dynamic time headway constraints among adjacent trains throughout the entire distance horizon. Moreover, a dynamic programming method is developed for comparative study, to verify the effectiveness of the proposed method.
|
|
| 6. |
- Du, Wei, et al.
(author)
-
Real-time Eco-Driving Control with Mode Switching Decisions for Electric Trucks with Dual Electric Machine Coupling Propulsion
- 2023
-
In: IEEE Transactions on Vehicular Technology. - 0018-9545 .- 1939-9359. ; 72:12, s. 15477-15490
-
Journal article (peer-reviewed)abstract
- This paper proposes a locally convergent, computationally efficient model predictive controller with mode switching decisions for the eco-driving problem of electric trucks. The problem is formulated as a bi-level program where the high-level optimises the speed trajectory and operation mode implicitly, while the low-level computes an explicit policy for power distribution of two electric machines. The alternating direction method of multipliers (ADMM) is employed at the high-level to obtain a locally optimal solution considering both speed optimisation and integer switching decisions. Simulation results indicate that the ADMM operates the powertrain with 0.9% higher total cost and 0.86% higher energy consumption than the global optimum obtained by dynamic programming for a quantised version of the same problem. Compared to a benchmark solution that maintains a constant velocity, the ADMM, running in a model predictive control framework (ADMM_MPC), operates the powertrain with a 8.77% lower total cost and 10.3% lower energy consumption, respectively. The average time for each ADMM_MPC update is 4.6ms on a standard PC, indicating its suitability for real-time control. Simulation results also show that the prediction errors of speed limits and road slope in ADMM_MPC cause only 0.12%-0.56% performance degradation.
|
|
| 7. |
- Du, Wei, et al.
(author)
-
Stochastic Model Predictive Energy Management of Electric Trucks in Connected Traffic
- 2023
-
In: IEEE Transactions on Vehicular Technology. - 0018-9545 .- 1939-9359. ; 72:4, s. 4294-4307
-
Journal article (peer-reviewed)abstract
- This paper proposes a cost-effective power management strategy utilizing the data provided by V2I communication techniques for dual electric machine coupling propulsion trucks. We formulate a bilevel program where the high-level optimizes operation mode implicitly, while the low-level computes an explicit policy for power distribution of two electric machines. Stochastic model predictive control (SMPC) strategy is employed at the high-level, the performance of which highly depends on the prediction accuracy of future driving information. To establish a position-dependent stochastic velocity predictor using limited amount of historical data, two improved approaches are developed: 1) Predictor using multiple features; 2) Predictor combining data and model. Simulations are performed to validate the performance of the proposed predictors compared with a benchmark. The results show that the controllers using the proposed predictors can reduce driving cost by 3.36 % and 4.26 %, respectively.
|
|
| 8. |
- Ebberstein, Victor, 1995, et al.
(author)
-
A unified framework for online trip destination prediction
- 2022
-
In: Machine Learning. - : Springer Science and Business Media LLC. - 0885-6125 .- 1573-0565. ; 111:10, s. 3839-3865
-
Journal article (peer-reviewed)abstract
- Trip destination prediction is an area of increasing importance in many applications such as trip planning, autonomous driving and electric vehicles. Even though this problem could be naturally addressed in an online learning paradigm where data is arriving in a sequential fashion, the majority of research has rather considered the offline setting. In this paper, we present a unified framework for trip destination prediction in an online setting, which is suitable for both online training and online prediction. For this purpose, we develop two clustering algorithms and integrate them within two online prediction models for this problem. We investigate the different configurations of clustering algorithms and prediction models on a real-world dataset. We demonstrate that both the clustering and the entire framework yield consistent results compared to the offline setting. Finally, we propose a novel regret metric for evaluating the entire online framework in comparison to its offline counterpart. This metric makes it possible to relate the source of erroneous predictions to either the clustering or the prediction model. Using this metric, we show that the proposed methods converge to a probability distribution resembling the true underlying distribution with a lower regret than all of the baselines.
|
|
| 9. |
- Egardt, Bo, 1950, et al.
(author)
-
Electromobility Studies Based on Convex Optimization DESIGN AND CONTROL ISSUES REGARDING VEHICLE ELECTRIFICATION
- 2014
-
In: IEEE Control Systems. - 1066-033X. ; 34:2, s. 32-49
-
Journal article (peer-reviewed)abstract
- This article presents a framework to study design tradeoffsin the search for electromobility solutions based on approximatemodeling of the power flows in the powertrain as afunction of component sizes. An important consequence ofthe modeling assumptions is that the optimal energy managementand component sizes can be computed simultaneouslyin a convex program, which means that competingdesigns can be evaluated in an objective way, avoiding theinfluence of a separate control system design. The fact thatthe optimization problem is convex allows large problemsto be solved with moderate computational resources, whichcan be exploited by, for example, running optimizationsover very long driving cycles. The problem formulationalso admits design decisions for the charging infrastructureto be included in the optimization.
|
|
| 10. |
- Elawad, Amal, 1989, et al.
(author)
-
Autonomous Bus Docking for Optimal Ride Comfort of Standing Passengers
- 2024
-
In: IEEE Transactions on Intelligent Transportation Systems. - 1524-9050 .- 1558-0016. ; In Press
-
Journal article (peer-reviewed)abstract
- This paper studies the optimization of ride comfort during the maneuver of bus docking at a stop station. We propose an analytical comfort model that considers the coupled and nonlinear effect of acceleration and jerk levels on the comfort perceived by standing bus passengers. This is studied through offline path planning by formulating the docking problem as a Nonlinear Program while implementing the comfort model to minimize discomfort. Geometry constraints are imposed on several points on the vehicle contour to ensure that the bus stays within the road bounds and docking is performed safely. The offline solution is then used as a reference in a real-time control event using a Volvo 7900 autonomous bus. To gain perspective, the measurements from the field test were compared to those of a human-driven trajectory. The results indicate that only around $\mathbf{\SI{1.7}{\%}}$ of bus occupants will experience discomfort with the proposed model, in comparison to $\mathbf{\SI{60}{\%}}$ in a human-driven bus, while respecting road and vehicle constraints and accurately docking within an acceptable predefined distance from the curb. We also show through simulations that in comparison to our proposed model, the traditional method of quadratic penalties for comfort modeling produces higher discomfort levels and prevents some trajectories characterized by high acceleration and jerk. Such trajectories can be permitted and comfortable using our approach.
|
|
