| 1. |
- Basso, Rafael, 1979, et al.
(author)
-
Dynamic Stochastic Electric Vehicle Routing with Safe Reinforcement Learning
- 2022
-
In: Transportation Research Part E: Logistics and Transportation Review. - : Elsevier BV. - 1366-5545. ; 157:157
-
Journal article (peer-reviewed)abstract
- Dynamic routing of electric commercial vehicles can be a challenging problem since besides the uncertainty of energy consumption there are also random customer requests. This paper introduces the Dynamic Stochastic Electric Vehicle Routing Problem (DS-EVRP). A Safe Reinforcement Learning method is proposed for solving the problem. The objective is to minimize expected energy consumption in a safe way, which means also minimizing the risk of battery depletion while en route by planning charging whenever necessary. The key idea is to learn offline about the stochastic customer requests and energy consumption using Monte Carlo simulations, to be able to plan the route predictively and safely online. The method is evaluated using simulations based on energy consumption data from a realistic traffic model for the city of Luxembourg and a high-fidelity vehicle model. The results indicate that it is possible to save energy at the same time maintaining reliability by planning the routes and charging in an anticipative way. The proposed method has the potential to improve transport operations with electric commercial vehicles capitalizing on their environmental benefits
|
|
| 2. |
- Basso, Rafael, 1979, et al.
(author)
-
Electric vehicle routing problem – a nested two level approach
- 2017
-
In: Proceedings of Swedish transportation research conference Stockholm 17-18 October 2017.
-
Conference paper (peer-reviewed)abstract
- Some of the main constraints of electric vehicles are related to their battery, in terms of energy capacity, time to recharge, weight and cost. One of the most important consequences is a limitation in driving range, which especially impacts commercial vehicles. Therefore in order to plan routes for this kind of vehicle, it is necessary to precisely estimate the energy required to drive and plan for charging whenever needed. This paper introduces the Two-stage Electric Vehicle Routing Problem (2sEVRP) that considers detailed information about the paths when estimating energy consumption and planning the routes. First, a method to calculate cost parameters for the road network is outlined including topography, speed, powertrain efficiency and the effect of acceleration and braking at traffic lights and intersections. Second, an integrated two-stage approach is described, which finds the best paths between pairs of nodes and then finds the best routes including battery and time-window constraints. Energy consumption is used as cost function including payload and auxiliary systems. The road cost parameters are aggregated to generate the path cost parameters that are used in the routing problem. In this way all the details of the paths are taken into account when computing energy demand for the routes. Last, numerical experiments were conducted with the road network from Gothenburg-Sweden and high-fidelity vehicle model simulations, focusing on trucks for urban distribution of goods. The results indicate that time and energy estimation are signeficantly more precise than existing methods. Consequently it is possible to generate important savings and be sure that the planned routes are feasible.
|
|
| 3. |
- Basso, Rafael, 1979, et al.
(author)
-
Electric Vehicle Routing Problem with Machine Learning for Energy Prediction
- 2021
-
In: Transportation Research Part B: Methodological. - : Elsevier BV. - 0191-2615. ; 145, s. 24-55
-
Journal article (peer-reviewed)abstract
- Routing electric commercial vehicles requires taking into account their limited driving range, which is affected by several uncertain factors such as traffic conditions. This paper presents the time-dependent Electric Vehicle Routing Problem with Chance- Constraints (EVRP-CC) and partial recharging. The routing method is divided into two stages, where the first finds the best paths and the second optimizes the routes. A probabilistic Bayesian machine learning approach is proposed for predicting the expected energy consumption and variance for the road links, paths and routes. Hence it is possible to consider the uncertainty in energy demand by planning charging within a confidence interval. The energy estimation is validated with data from electric buses driving a public transport route in Gothenburg-Sweden as well as with realistic simulations for 24 hours traffic in the city of Luxembourg connected to a high fidelity vehicle model. Routing solutions are compared with a deterministic formulation of the problem similar to the ones found in the literature. The results indicate high accuracy for the energy prediction as well as energy savings and more reliability for the routes.
|
|
| 4. |
- Basso, Rafael, 1979, et al.
(author)
-
Energy consumption estimation integrated into the Electric Vehicle Routing Problem
- 2019
-
In: Transportation Research Part D: Transport and Environment. - : Elsevier BV. - 1361-9209. ; 69, s. 141-167
-
Journal article (peer-reviewed)abstract
- When planning routes for fleets of electric commercial vehicles, it is necessary to precisely predict the energy required to drive and plan for charging whenever needed, in order to manage their driving range limitations. Although there are several energy estimation models available in the literature, so far integration with Vehicle Routing Problems has been limited and without demonstrated accuracy. This paper introduces the Two-stage Electric Vehicle Routing Problem (2sEVRP) that incorporates improved energy consumption estimation by considering detailed topography and speed profiles. First, a method to calculate energy cost coefficients for the road network is outlined. Since the driving cycle is unknown, the model generates an approximation based on a linear function of mass, as the latter is only determined while routing. These coefficients embed information about topography, speed, powertrain efficiency and the effect of acceleration and braking at traffic lights and intersections. Secondly, an integrated two-stage approach is described, which finds the best paths between pairs of destinations and then finds the best routes including battery and time-window constraints. Energy consumption is used as objective function including payload and auxiliary systems. The road cost coefficients are aggregated to generate the path cost coefficients that are used in the routing problem. In this way it is possible to get a proper approximation of the complete driving cycle for the routes and accurate energy consumption estimation. Lastly, numerical experiments are shown based on the road network from Gothenburg-Sweden. Energy estimation is compared with real consumption data from an all-electric bus from a public transport route and with high-fidelity vehicle simulations. Routing experiments focus on trucks for urban distribution of goods. The results indicate that time and energy estimations are significantly more precise than existing methods. Consequently the planned routes are expected to be feasible in terms of energy demand and that charging stops are properly included when necessary.
|
|
| 5. |
- Basso, Rafael, 1979
(author)
-
Energy consumption prediction and routing for electric commercial vehicles
- 2021
-
Doctoral thesis (other academic/artistic)abstract
- With the recent growing interest for electric vehicles as one of the initiatives to help tackle pollution and climate change, several opportunities and challenges emerge. This kind of vehicle releases no tailpipe emissions, is quieter, more energy efficient in terms of tank-to-wheels and simpler, which can lead to less maintenance. On the other hand, their battery is still the main limitation in terms of energy capacity, time to recharge, weight and cost. One of the main consequences is a limitation in driving range, which especially affects commercial vehicles. In order to adopt electric trucks for urban distribution of goods, there is a need to improve and adapt current planning tools to take into account their constraints. To plan the routes and charging for these vehicles it is necessary to estimate their energy consumption accurately. This thesis focuses on the development of energy consumption prediction and routing methods for electric commercial vehicles. The first part presents an overall background and short state of the art review. The main contributions are presented in the second part. The included articles are a step-by-step development of the methods, each covering different aspects of the problem. The first paper presents a deterministic energy prediction model integrated into routing models. The second paper proposes a probabilistic energy estimation method based on Bayesian machine learning and adds chance-constraints into the routing problem in order to plan charging within a confidence interval. The third paper covers routing with dynamic customers and stochastic energy consumption, proposing a solution method based on Safe Reinforcement Learning to minimize the risk of battery depletion by planning charging in an anticipative way. All papers are validated with realistic simulations as well as logged data. The results indicate that it is possible to save energy and reduce the risk of running out of energy while en route.
|
|
| 6. |
- Basso, Rafael, 1979
(author)
-
Route planning and energy consumption estimation for electric commercial vehicles
- 2017
-
Licentiate thesis (other academic/artistic)abstract
- With the recent growing interest for electric vehicles as one of the initiatives to help tackle pollution and climate change, several opportunities and challenges emerge. This kind of vehicle releases no tailpipe emissions, is quieter, more energy efficient in terms of tank-to-wheels and simpler, which can lead to less maintenance. On the other hand their battery is still the main limitation in terms of energy capacity, time to recharge, weight and cost. One of the main consequences is a limitation in driving range, which especially impacts commercial vehicles. In order to adopt electric trucks for urban distribution of goods, there is a need to improve and adapt current planning tools to take into account their constraints. To plan the routes and charging for these vehicles it is necessary to precisely estimate their energy consumption.This thesis gives an overall background and state of the art review in the introductory chapters. The main contributions are presented in the second part. The first paper describes a time-dependent electric vehicle routing problem. It also analyses the different factors that affect energy consumption and routing for electric vehicles. The second paper introduces the Two-stage Electric Vehicle Routing Problem (2sEVRP), with a precise energy consumption estimation model, a first stage to find the best paths between the nodes to be visited and the second stage to find the route considering time-windows and planning charging when necessary. The paper shows numerical experiments with the road network from Gothenburg-Sweden and a high-fidelity vehicle simulation. The results indicate higher precision in energy estimation and savings while routing when comparing to existing approaches from the literature.
|
|
| 7. |
- Basso, Rafael, 1979, et al.
(author)
-
Traffic aware electric vehicle routing
- 2016
-
In: IEEE Conference on Intelligent Transportation Systems, Rio de Janeiro, Brazil,November 1-4. ; , s. Art no 7795588, Pages 416-421
-
Conference paper (peer-reviewed)abstract
- Since the main constraint of electric vehicles is range due to limited battery capacity, the focus for routing these kind of vehicles should be energy consumption minimization. And since energy consumption depends on several aspects, this article introduces a new model for route optimization of Electric Commercial Vehicles, with a realistic energy consumption model based on factors such as road inclination, weight and speed. The main new feature is to consider average speed for the road network at different times during the day, with the vehicle adapting to traffic flow. Several experiments were performed to evaluate the impact of different elements in energy consumption. As a result a few topics are recommended for future work.
|
|
| 8. |
- Bruzelius, Fredrik, et al.
(author)
-
Test report
- 2010
-
Reports (other academic/artistic)abstract
- This report is the document summarising the testing experience and knowledge gained from the physical testing activities within the eVALUE project. The document contains brief introductions to the testing scenarios as well as short summaries of conclusions. Appended to this document are the testing reports that have been compiled during the different test sessions. Physical testing at test tracks all across Europe has been the main input in the development of scenarios and test procedures. This document describes the development tests that have been performed during 2010, based on a first draft set of testing protocols. The experience from the performed tests has been used as an important input to the revision of the testing protocols, i.e. the formal documents that describe how a test should be performed and evaluated. These protocols are documented in the separate Deliverable 3.2.
|
|
| 9. |
- Jacobson, Jan, et al.
(author)
-
Final testing protocols
- 2011
-
Reports (other academic/artistic)abstract
- This report is the final document summarizing the inspection and testing protocols of the eVALUE project. It describes principles, inspection protocols and testing protocols for performance testing of ICT-based safety systems. The inspection protocols (published earlier in D2.2) and the testing protocols introduced in D3.1 are replaced by the ones in D3.2. The older versions are obsolete and should be disregarded. The inspection protocols cover the definition of the test vehicle, HMI aspects, environmental conditions, and functional safety. The inspection protocols are used to prepare for the physical tests as well as evaluating the performance of the vehicle. The testing protocols address longitudinal, lateral, and stability-oriented traffic scenarios. The longitudinal scenarios include a pedestrian crossing the road in front of the vehicle, or the situation where a driver approaches a stationary queue of cars. Involuntarily lane departures and cars in the blind spot during a lane change are situations covered by the lateral scenarios. Exiting a highway, avoiding an obstacle, and braking on a partially ice-covered road surface are examples of traffic scenarios related to stability.
|
|
| 10. |
- Sanchez-Diaz, Ivan, 1984, et al.
(author)
-
A Time-Efficiency Study of Medium-Duty Trucks Delivering in Urban Environments
- 2020
-
In: Sustainability. - 2071-1050. ; 12:1, s. 425-
-
Journal article (peer-reviewed)abstract
- This paper uses data from a major logistics service provider in Gothenburg (Sweden) to (i) identify the different activities in a typical urban distribution tour, (ii) quantify the time required by drivers to perform each of these activities, and (iii) identify potential initiatives to improve time efficiency. To do so, the authors collected GPS data, conducted a time-study of the activities performed by the drivers for a week, conducted a focus group with the drivers, and a set of interviews with managers. The results show that driving represents only 30% of the time, another 15% is spent on breaks, and the remaining 55% is used to perform activities related to customer service, freight handling, and planning. The latter are subdivided into multiple activities, each taking a small amount of time. A focus group with the drivers and some interviews revealed several initiatives to improve time efficiency. Most initiatives can bring small gains, but when aggregating all potential time savings there is a big potential to improve overall time efficiency. Initiatives with highest potential and low cost are: providing better pre-advice on upcoming customers, improving route planning, having hand-free cell phone use, and enhancing handling equipment.
|
|
