Energy management control system

Abstract

Determining a charging schedule for a battery of an electric drive system of a vehicle comprises obtaining a forecast period 402, determining a predicted driving schedule comprising a plurality of pre

Description

TECHNICAL FIELD The present invention relates to a computer implemented method of determining a charging schedule for a vehicle. Aspects of the invention relate to a method, to a control system, and to a vehicle. BACKGROUND TO THE INVENTION Although electric vehicles (EVs) do not produce tailpipe emissions directly, the charging stations used to recharge the batteries of an EV or a hybrid electric vehicle are typically connected to an electric grid or power network, which is powered by non-renewable energy sources or a mix of renewable and non-renewable energy sources. These non-renewable energy sources, such as fuel-burning power plants, produce large-scale emissions that impact the total environmental impact associated with the EV during its operating lifecycle. Carbon intensity data is available for comparing the emissions associated with respective charging stations and power sources, such that users can choose to charge their vehicle at a particular charging station in order to minimise the associated environmental impact. However, users are likely to miss the “best” (i.e., least carbon intense) times and locations for charging the EV during ordinary driving of the vehicle. It is against this background that the invention has been devised. SUMMARY OF THE INVENTION According to an aspect of the invention, there is provided a computer implemented method of determining a charging schedule for (a battery of) an electric drive system of a vehicle. The method comprises obtaining a forecast period for a charging schedule. The method further comprises determining a predicted driving schedule comprising a plurality of predicted journeys for the forecast period using a behavioural model configured to model driving behaviour of the vehicle. The method further comprises: determining a predicted journey area based on the plurality of predicted journeys; identifying one or more charging locations located within the predicted journey area; and determining predicted emissions data associated with the one or more charging locations for the forecast period. The method also comprises determining the charging schedule based on the predicted driving schedule, the identified one or more charging locations, and the predicted emissions data associated with the one or more charging locations. The determined charging schedule comprises one or more suggested times and locations for charging the vehicle during the forecast period. The method is therefore able to provide charging recommendations in the form of a charging schedule by leveraging analysis of vehicle driving behaviour, as well as emission intensity data, in order to minimise the emissions associated with charging the vehicle during a forecast period. Ultimately, the method will lead to reduced emissions associated with charging and operating the vehicle. The vehicle may include an electric drive system, such as an electric vehicle or a hybrid vehicle. For each predicted journey, the predicted driving schedule may include information corresponding to coordinates, time, and / or date for each of the start and end of the predicted journey, and / or the route between the start and end of the predicted journey. The forecast period may be a reference period, ora period derived for a reference driving range. The forecast period may be user definable. The predicted journey area may be a bounding geographical area that encompasses the predicted journeys that the vehicle is expected to complete during the forecast period. The behavioural model may be a statistical model calibrated based on reference journey data indicative of the driving behaviour of the vehicle. Alternatively, the behavioural model may be a machine learning model trained based on reference journey data indicative of the driving behaviour of the vehicle. The reference journey data may, for example, comprise historic journey data indicative of a plurality of historic journeys completed by the vehicle. This beneficially allows for the driving schedule (and therefore the charging schedule) to consider actual behaviour of the user, as data collected during previous journeys made by the user. The historic journey data may be indicative of the date, time, and routes of each historic journey. In one example, the historic journey data may comprise one or more of the following for each historic journey: a start location and an end location; a start time and an end time; a route between the start and end locations; and / or a state of charge (SoC) of the vehicle and a predicted remaining range of the vehicle at the start and end location. This allows for the behavioural model to learn the general locations that the vehicle visits. The charging schedule is therefore able to be tailored to create a more bespoke charging schedule that fits user habits. Historic journey data may further comprise one or more of the following for each historic journey: the user’s preferred location for charging the electric vehicle, the user’s preferred time of day for charging the electric vehicle, the user’s preferred length of time for charging the electric vehicle. For each historic journey, historic journey data may further comprise other relevant data regarding locations of interest that may be visited between the start and end locations, such as brief visits to a store. In an example, the method may further comprise obtaining the historic journey data during a calibration stage. The method may further still comprise training the behavioural model based on the historic journey data prior to determining the predicted driving schedule. The method may further comprise receiving reference journey data via a user input device of the vehicle. This allows for the ability to generate a charging schedule without first needing to carry out any journeys in the vehicle. Determining the predicted emissions data may comprise applying one or more time series forecasting methods to historic emissions data for each of the identified charging locations. The historic emission data in some examples may comprise time series data of the power source and the associated emissions. The predicted emissions data may comprise a time series forecast of associated emissions for each power source during the forecast period. Alternatively, determining the predicted emissions data may comprise receiving the predicted emissions data from a third-party server. This beneficially provides the ability to predict low emission charging times based on historic low emission charging times. The method may further comprise outputting the charging schedule to a user of the vehicle. This allows the user to view and choose routes / charging locations to reduce emissions based on knowledge that they have about their upcoming schedules that may not be able to be predicted by existing / historic data. In an embodiment, determining the charging schedule may comprise determining which of the identified charging locations is located within a threshold proximity of the vehicle according to each predicted journey. Determining the charging schedule may further comprise determining an estimated time of intersection of the vehicle with that charging location. Determining the charging schedule may also comprise determining the predicted emissions associated with that charging location at the estimated time of intersection. The obtained forecast period may be updated at a prescribed frequency such that the charging schedule is redetermined for the updated forecast period. The current state of the vehicle may include current state of charge and current coordinates / location of the vehicle. The method may further comprise obtaining data indicative of a range of the vehicle. The charging schedule may be determined based, at least in part, on the indicated range of the vehicle. According to a further aspect of the invention, there is provided a computer program product, comprising computer readable instructions which, when the program is executed by one or more processors cause the one or more processors to perform the method described in a previous aspect of the invention. According to yet another aspect of the invention, there is provided a computer readable medium comprising instructions which, when executed by a computer, cause the computer to perform the steps of the method described in a previous aspect of the invention. According to another aspect of the invention there is provided an energy management control system for a vehicle. The energy management control system comprises an electric drive system and a battery for the electric drive system. The control system is configured to determine a charging schedule for the battery. The control system comprises one or more controllers configured to: obtain a forecast period for the charging schedule; determine a predicted driving schedule comprising a plurality of predicted journeys for the forecast period using a behavioural model configured to model driving behaviour of the vehicle; determine a predicted journey area based on the plurality of predicted journeys; identify one or more charging locations located within the predicted journey area; determine predicted emissions data associated with the one or more charging locations for the forecast period. The one or more controllers is also configured to determine the charging schedule based on: the predicted driving schedule; the identified one or more charging locations; and the predicted emissions data associated with the one or more charging locations. The determined charging schedule comprises one or more suggested times and locations for charging the vehicle during the forecast period. According to another aspect of the invention there is provided a vehicle comprising an energy management control system as described in a previous aspect of the invention. Within the scope of this application it is expressly intended that the various aspects, embodiments, examples and alternatives set out in the preceding paragraphs, in the claims and / or in the following description and drawings, and in particular the individual features thereof, may be taken independently or in any combination. That is, all embodiments and / or features of any embodiment can be combined in any way and / or combination, unless such features are incompatible. The applicant reserves the right to change any originally filed claim or file any new claim accordingly, including the right to amend any originally filed claim to depend from and / or incorporate any feature of any other claim although not originally claimed in that manner. BRIEF DESCRIPTION OF THE DRAWINGS One or more embodiments of the invention will now be described, by way of example only, with reference to the accompanying drawings, in which: Figure 1 is a schematic illustration of a plan view of a vehicle, in accordance with an embodiment of the invention, with an overlay of an exemplary energy management control system; Figure 2 schematically illustrates an exemplary embodiment of the control system shown in Figure 1; Figure 3 is a schematic illustration of an exemplary geographical map, which includes a plurality of exemplary predicted journeys, an exemplary predicted journey area and a plurality of charging locations identified within that predicted journey area; and Figure 4 schematically illustrates an exemplary method of determining a charging schedule for a vehicle, implemented by the control system shown in Figure 2. DETAILED DESCRIPTION Embodiments of the present invention relate to methods and corresponding control systems for energy management of a vehicle, specifically energy management of a vehicle with an electric drive system and battery energy storage system, such as an electric vehicle or a hybrid electric vehicle. The invention provides a system for determining charging recommendations in the form of a charging schedule, which leverages analysis of driving behaviour and emission intensity data to minimise the emissions associated with charging the vehicle during a forecast period. In particular, the driving behaviours and routines of a user / vehicle are modelled using one more statistical analysis methods and / or machine learning models for predicting a driving schedule based on reference journey data. The predicted driving schedule includes a plurality of predicted journeys that the vehicle is expected to complete during the forecast period, which may be a period of days, weeks, or months, for example, as selected or configured. The range of predicted journeys are used to determine a wider predicted journey area within which the vehicle travels and to identify charging location(s) located within the predicted journey area that could be used to conveniently charge the vehicle. In this way, potential charging location recommendations are limited to those charging location(s) that the user is likely to encounter or approach according to their predicted driving behaviour. The predicted journey area shall therefore encompass the plurality of predicted journeys and may, for example, be determined using one or more schemes, rules, or functions. For example, the predicted journey area may be defined so as to encompass areas within a threshold distance, radius, or proximity of each predicted journey, and / or by fitting (i.e. locating and sizing) an envelope of a predetermined shape for encompassing all of the predicted journeys. Predicted emissions data is obtained (e.g. from a third-party) or otherwise determined for the forecast period for each of the identified charging location(s). For example, for each charging location, the predicted emissions data may be determined using one or more time series forecasting methods based on historic emissions data associated with that charging location. In each case, the precited emissions data provides an estimate of the time varying emissions expected to be produced by the anticipated sources of energy for the charging location. The predicted driving schedule, the identified charging location(s), and the predicted emissions data are then combined by the method / system to determine one or more suggested times and locations for charging the vehicle during the forecast period, forming a charging schedule. For example, the charging schedule may be determined to achieve one or more charging objectives, such as minimising overall emissions, which may be preprogrammed by a manufacturer, for example. The suggested charging times and locations may be determined accordingly based on a comparative assessment. The determined charging schedule may be put to various uses and may be output to a user as a suggestion, via a human-machine interface, or used to update vehicle navigation. It is envisaged that the method / system of the invention will lead to enhanced utilisation of renewable and low-emission energy sources for charging electrically driven vehicles, ultimately leading to reduced emissions associated with operating the EVs. Embodiments of the invention shall now be discussed in more detail with reference to Figures 1 to 4. Figure 1 shows a plan view of a vehicle 100, in accordance with an embodiment of the invention, with a schematic overlay of an energy management control system 102, for managing the energy usage of the vehicle 100. In this example, the vehicle 100 takes the form of a passenger car. However, this example is not intended to be limiting on the scope of the invention and, in other examples, the vehicle 100 may take various other forms, such as a bus, or truck, amongst other passenger vehicles. Moreover, whilst the vehicle 100 is shown to include an electric powertrain 104 in Figure 1, it shall be appreciated that the vehicle 100 may alternatively include a hybrid powertrain arrangement, i.e. including both internal combustion and electric drive systems. The electric powertrain 104 includes one or more electric machines or electric drive motors 105 and a battery energy storage system or battery 106, such as a high-voltage (HV) battery. The battery 106 may be a rechargeable battery, also known as a traction battery, having a high power-to-weight ratio and energy density. For example, the battery 106 may take the form of a lithium-ion battery or battery pack. The battery 106 can be charged and recharged at a charging station, or charging location, by connection to an electric grid or power network. Although not shown here, the battery 106 may also be recharged by one or more energy recovery systems of the vehicle 100, such as a regenerative braking system. The battery 106 provides the source of electrical power for the electric drive motors 105. At any given time, the battery 106 has a state of charge, for example between 0% and 100%, signifying the remaining capacity of the battery 106 as a percentage of a maximum battery capacity. The state of charge of the battery 106 may be used for estimating a remaining range or mileage of the vehicle 100 according to one or more known methods. It shall be appreciated that the electric powertrain 104 may therefore further include a battery management system (not shown) for managing the battery 106. For example, the battery management system may be configured to monitor the state of charge and state of health of the battery 106, and / or control the charging and discharging of the battery 106. The energy management control system 102 is configured to determine a charging schedule for the vehicle 100, ora particular user of the vehicle 100, for a forecast period. The determined charging schedule includes one or more suggested times and locations for charging the vehicle during the forecast period. The suggested time(s) and location(s) are determined so as to balance a vehicle I user’s driving behaviour and routine against emission or carbon intensity data associated with different charging locations. In this manner, the determined charging schedule recommends optimum charging locations and times for minimising the emissions associated with charging the vehicle 100 during a predicted driving routine. As shown in Figure 1, the energy management control system 102 may also connect to one or more additional systems 108 of the vehicle 100 for communicating the determined charging schedule, or individual suggestions thereof, to the user. To give an example, the additional system(s) 108 may include a communication system 109 for communicating the determined charging schedule to the user. The communication system 109 may, for example, include or connect to an audio system of the vehicle 100, and / or a humanmachine interface (HMI) device or user interface 120, as shown in Figure 1. The user interface 120 may also be configured to enable the user to provide one or more inputs to the control system 102. For example, the user interface 120 may be configured to enable the user to provide inputs relating to the forecast period, and / or charging objectives, amongst other inputs, as will be explained further below. In some embodiments, the user interface 120 may include a display and one or more input devices, such as touchscreens, keypads, buttons and / or the like, which are configured to receive inputs from the user. In other embodiments, the user interface 120 may take the form of a touchscreen mounted on a dashboard of the vehicle 100. In still further embodiments, the user interface 120 may also be a smartphone, tablet, or other such user device, connected to the control system 102. Alternatively, or additionally, the energy management control system 102 may connect to the one or more additional systems 108 for controlling the vehicle 100 according to the determined charging schedule. An exemplary energy management control system 102, in accordance with embodiments of the present invention, shall now be discussed in more detail with additional reference to Figure 2. As shown in Figure 2, the energy management control system 102 includes a journey forecasting system 114, an emissions forecasting system 116, a charge scheduling system 118, and a memory 122, in this example. That is, the control system 102 is shown to include four functional elements, units, modules, or sub-systems in Figure 2. Each of these units, modules or sub-systems may be provided, at least in part, by suitable software running on any suitable computing substrate using conventional or custom processors and memory. Some or all of the units or modules may use a common computing substrate (for example, they may run on the same server) or separate substrates, or different combinations of the modules may be distributed between multiple computing devices. The example architecture of the control system 102 is not intending to be limiting on the scope of the invention though and, in other examples, it shall be appreciated that the architecture may take other suitable forms. Each of the journey forecasting system 114, the emissions forecasting system 116, and the charge scheduling system 118 may comprise at least one processor connected to the memory 122. The processors may be configured to execute machine-readable code or instructions stored in the memory 122 or received from other computer readable media (e.g. CDROM, network storage, a remote server, etc.). In some embodiments, the memory 122 may include one or more devices (e.g. memory units, memory devices, storage devices, or other computer-readable media) for storing data and / or computer code for completing and / or facilitating the various processes described in the present disclosure. The memory 122 may include random access memory (RAM), read-only memory (ROM), hard drive storage, temporary storage, non-volatile memory, flask memory, optical memory, or any other suitable memory for storing software object and / or computer instructions. The memory 122 may be communicatively connected to each of the processors via a processing circuit and may include computer code for executing (e.g., by the processors) one or more of the processes described herein. The journey forecasting system 114 is configured to determine a predicted driving schedule and a corresponding predicted journey area for a forecast period. The forecast period is a reference time period, which may be a length of time (such as one or more days, weeks, months, etc, from the current point in time) or a time period bound by specified dates, such as 1 March 2024 to 10 March 2024 for example. A user may enter the reference time period (as an input) via the user interface 120 connected to the control system 102, and the journey forecasting system 114 may subsequently access this reference time period for use in determining the charging schedule. Alternatively, or additionally, the forecast period may be pre-programmed and stored in the memory 122. For example, the forecast period may be set to a predetermined interval period and updated iteratively, for example at a predetermined frequency or in response to further user inputs. The journey forecasting system 114 is configured to use the forecast period as the input parameter for predicting a corresponding driving schedule. The predicted driving schedule includes a plurality of predicted journeys that a vehicle / user is expected to complete during the forecast period. For each predicted journey, the predicted driving schedule may include information corresponding to coordinates, time, and / or date for each of the start and end points of the predicted journey, and / or the predicted route between the start and end of the predicted journey. For this purpose, it shall be appreciated that the journey forecasting system 114 may include or have access to geographical information, such as a map. For example, such geographical information may be stored in the memory 122 or the control system 102 may access such information via a vehicle navigation system 124 connected to the control system 102, as shown in Figure 2. For example, the navigation system 124 may form one of the additional system(s) 108 shown in Figure 1. The navigation system 124 may take the form of, or otherwise include, a global navigation satellite system (GNSS), for example, storing such geographical information. The control system 102 may therefore interact with the navigation system 124 to access the geographic information and / or to determine spatiotemporal characteristics of the vehicle 100, i.e. the geographic location of the vehicle 100 at a particular point in time. By way of example, Figure 3 shows a portion of an exemplary geographical map 300, which includes a plurality of predicted journeys 301 a-d that a vehicle / user is expected to complete during a forecast period. In this simplified example, the forecast period correspond to ‘tomorrow’ and it shall be appreciated that the forecast period may be updated in this manner on a daily basis for the next day. The plurality of predicated journeys 301 a-d is shown to include first, second, third and fourth predicted journeys 301a-d, in this example, each being depicted by a dashed line. The first predicted journey 301a extends from a home location 304 of the user / vehicle to a first destination 305, for example with a predicted start time of 07:30 and a predicted arrival time of 07:55. The second predicted journey 301b extends from the first destination 305 to a second destination 307, for example with a predicted start time of 08:15 and a predicted arrival time of 08:25. The third predicted journey 301c extends from the second destination 307 to a work location 306 of the user, for example with a predicted start time of 08:40 and a predicted arrival time of 09:00. Lastly, the fourth predicted journey 301d extends from the work location 306 back to the home location 304, for example at the end of the workday with a predicted start time of 17:15 and a predicted arrival time of 18:00. In order to determine the predicted driving schedule, the journey forecasting system 114 includes a behavioural model 114a which is trained, validated, and / or configured using reference journey data associated with a particular subject. For example, the reference journey data may be associated with the vehicle 100 or a particular user of the vehicle 100 (e.g. with individual users being associated with respective user profiles stored in the memory 122). The behavioural model 114a is configured to model the driving behaviours and routines of that subject for the purpose of predicting future journeys. For example, the behavioural model 114a may take the form of a machine learning model, including one or more machine learning algorithms that are trained to predict journeys based on reference journey data associated with that subject. The behavioural model 114a may therefore comprise an artificial neural network, a deep neural network, or similar. To give an example, the behavioural model 114a may take the form of a recurrent neural network RNN, for example of the long-short term memory type (able to process sequential data and capture long-term dependencies in the driving behaviour of the subject). In this manner, the behavioural model 114a may act as a classification model trained to estimate the probability of the subject taking different journeys at respective dates / times. It shall be appreciated that one or more training methods that are known in the art may be used for training such a neural network to predict the driving schedules for a particular subject. Such methods may include the use of one or more random decision forests and / or a logistic regression analysis, for example, amongst other techniques. The reference journey data used to train the behavioural model 114a may be obtained by the journey forecasting system 114 according to one or more methods. For instance, the reference journey data may include data which is indicative of historic journeys (i.e. previous journeys) that the subject has completed, i.e. historic journey data. The historic journey data may therefore be collected by the navigation system 124 of the vehicle 100, for example. In an example, the historic journey data may be stored in the memory 122 of the control system 102 for subsequent access by the journey forecasting system 114. For example, the vehicle navigation system 124 may be configured to monitor the route of the vehicle 100 by determining the current vehicle position in a reference co-ordinate system during driving between start and end points. The connection to the vehicle navigation system 124 therefore facilitates the collection of data related to historic journeys completed by a vehicle. The historic journey data may include information corresponding to coordinates, time, and / or dates for each of the start and end points of each historic journey, and / or the route taken between the start and end of the historic journey. For example, for each historic journey, the obtained data may comprise one or more of the following: a start location and an end location, a start time and an end time, a route between the start and end locations, a state of charge (SoC) of the vehicle 100 (i.e. level of the battery 106) and a predicted remaining range of the vehicle 100 at the start and / or end locations. It shall be appreciated that such reference journey data may be pre-processed to clean and normalise the data for use in training the behavioural model 114a. In examples, the historic journey data may be collected during an initial “calibration stage” of the control system 102. That is, the control system 102 may be configured to execute an initial calibration stage, e.g. upon initial activation of the control system 102, for the collection of reference journey data for use in configuring the behavioural model 114a. It shall be appreciated that the reference journey data should therefore be indicative of the driving behaviours, routines, and patterns, of the subject such that future journeys can be predicted with accuracy by suitable extrapolation or knowledge transfer. For example, the calibration stage may record journey data for a prescribed period or number of journeys to acquire sufficient data for obtaining an acceptable modelling accuracy. To give an example, the calibration stage may last for a minimum period of one month, during which journey data is recorded and stored as historic journey data in the memory 122 for subsequent use in configuring the behavioural model 114a. In other examples, the calibration stage may last until the behavioural model 114a has been validated, e.g. by determining predicted journeys with a threshold accuracy level. It shall be appreciated that the forecasting system 114 may be configured to validate the behavioural model 114a using one or more modelling validation methods that are known in the art, which shall not be described in detail here to avoid obscuring the invention. The reference journey data may also include information indicative of one or more user preferences, such as identified preferred geographic locations, times of day for driving or charging, and lengths of time for charging the electric vehicle 100. Preferred geographic locations are locations of significance to the user, which may be visited frequently and may be identified by the journey forecasting system 114 based on repeat visits of the vehicle 100 to these locations (for example exceeding a prescribed threshold). Preferred geographic locations may therefore include home and work locations, for example. Additionally, or alternatively, preferred geographic locations may also include frequently visited shops, cafes, restaurants, homes of friends and family, etc. In examples, the calibration stage may be initiated, paused, and / or terminated by one or more user inputs, and the control system 102 may then proceed to an active stage, during which the behavioural model 104 is used to determine predicted driving schedules for respective forecast periods. During the active stage, it shall be appreciated that the control system 102 may be configured to continue recording journey data and refining the behavioural model 114a based thereon. In this manner, the behavioural model 114a may be continuously or periodically updated for improved robustness and adaptation to driving behaviour changes. In other examples, a user may enter reference journey data into the user interface 120, and the journey forecasting system 114 may subsequently access this reference journey data from the user interface 120. Additionally, the invention is not limited to the use of machine learning models for predicting the driving schedule, and in other examples, the behavioural model 114a may take the form of a statistical model, for example, comprising one or more algorithms or functions for predicting vehicle journeys based on the reference journey data. In such examples, the model may therefore include one or more statistical functions with parameters derived using statistical analysis of the reference journey data. Such statistical analysis may involve the use of time series forecasting techniques, such as simple moving average, exponential smoothing, autoregressive integration moving average, or Croston modelling, as would be appreciated by the skilled person. The journey forecasting system 114 is further configured to determine the predicted journey area 302 based on the plurality of predicted journeys 301 a-d. For this purpose, the journey forecasting system 114 may include one or more rules, schemes, or algorithms, for defining an area that encompasses the plurality of predicted journeys 301 a-d, along with nearby charging locations. For example, the journey forecasting system 114 may be configured to identify a bounding area or envelope encompassing the plurality of predicted journeys and areas within a threshold distance or proximity of each predicted journey. Alternatively, or additionally, the journey forecasting system 114 may be configured to determine the predicted journey area 302 by fitting (i.e., sizing and locating) a boundary / envelope of a predetermined shape that encompasses the plurality of predicted journeys, e.g. using one or more fitting functions. As shown in Figure 3, the predicted journey area 302 therefore takes the form of a bounding geographical area that encompasses the predicted journeys 301 a-d (i.e., the start and end locations and predicted routes) that the vehicle 100 is expected to complete during the forecast period. The journey forecasting system 114 is also configured to identify those charging locations that are located within the predicted journey area 302. For this purpose, the journey forecasting system 114 may include or have access to a database of predetermined charging locations, such as publicly available charging points and charging stations. This charging location information indicates the geographic location of each charging location and may be obtained or derived from the same geographic information or map used to determine the routes and / or journey area. The journey forecasting system 114 may then identify each charging location located within the predicted journey area 302. The predicted journey area 302 and identified charging locations may be shown on a map and displayed to the user via the user interface 120. For example, the geographical map 300 shown in Figure 3 shows a plurality of public charging stations 308, 310, 312 located within the predicted journey area 302. In other examples, home locations, work locations, and other private charging locations may also be identified by the journey forecasting system 114, as shall be described in more detail. The emissions forecasting system 116 is configured to determine predicted emissions data associated with individual charging locations for respective forecast periods. In particular, the emissions forecasting system 116 is configured to determine predicted emissions data associated with each of the identified charging locations 308 - 312 during the forecast period. Similarly to the journey forecasting system 114, the emissions forecasting system 116 may include or have access to a database of power emissions information, including information indicating the geographic location of each charging location, and information relating to the emissions associated with the power supplied to that charging location. For example, the information relating to the emissions associated with each charging location may include data indicative of the power source supplying each charging station 308-312 and the associated emissions (such as g-CO2 per kWh) at a particular time and date. The database may include predicted emissions and the emissions forecasting system 116 may be configured to lookup the predicted emissions data for each of the identified charging locations 308-312 during the forecast period. Alternatively, the database may be limited to measured emissions data and the emissions forecasting system 116 may be configured to determine predicted emissions data for each of the identified charging locations 308-312 based on current and historic emissions data. The historic emissions data may be iteratively updated, for example at a prescribed frequency, to encompass more recent data. In each case, the emissions forecasting system 116 may be configured to access the database or the emission data via a third-party server. Alternatively, the database of information may be stored in the memory 122 and accessed by both the journey forecasting system 114 and the emissions forecasting system 116. In order to determine the predicted emissions data, the emissions forecasting system 116 may be configured to apply one or more time series forecasting methods to the historic emissions data for each of the identified charging locations 308-312. The predicted emissions data may therefore comprise a time series forecast of associated emissions (i.e., CO2 emissions) for each power source, and therefore for each charging location 308-312, during the forecast period. The charge scheduling system 118 is configured to determine the charging schedule for the vehicle 100 based on the predicted driving schedule, the identified charging locations, and the predicted emissions data associated with the identified charging locations. The charging schedule includes at least one recommended charge time and corresponding location for the forecast period. In examples, the charge scheduling system 118 may also be configured to determine the charging schedule based on a determined charging frequency of the vehicle 100. Here, the charging frequency relates to the number of times that the vehicle 100 is charged by the user (i.e., a “charging event”) within the forecast period. A charging event occurs when the vehicle 100 is plugged in to a charging location for a threshold period of time or amount of charge delivered to the battery 106 (i.e., an increase in state of charge of the vehicle 100 that exceeds a minimum threshold). For example, if the forecast period is defined as one week, the charging frequency may be three charging events per week. The charge scheduling system 118 may be configured to determine the charging frequency from the reference journey data. More specifically, the charge scheduling system 118 may be configured to determine the charging frequency from the historic journey data, i.e. an algorithm or other means may be used by the charge scheduling system 118 to determine how often the user charges the vehicle 100 in a given time period based on the gathered journey data. Alternatively, a user may enter this charging frequency via the user interface 120 connected to the control system 102, and the charge scheduling system 118 may subsequently access this charging frequency for use in determining the charging schedule. Alternatively, or additionally, the charging frequency may be pre-programmed and stored in the memory 122. For example, the charging frequency may be set to a predetermined frequency and updated iteratively, for example at a predetermined frequency or in response to further user inputs. In each case, the charge scheduling system 118 may determine the charging schedule based, at least in part, on the determined charging frequency, for example to determine a frequency of charge recommendation. That is, If the charging frequency is a factor in determining the charging schedule, then the charging schedule may include recommended charge time(s) and corresponding location(s) for the forecast period that correlate to the charging frequency, e.g. for a charging frequency of three charging events per forecast period, the charging schedule may include three charging times and corresponding locations for the forecast period. The charge scheduling system 118 may be configured to determine the charging schedule according to one or more charging objectives, such as minimising emissions associated with charging the vehicle 100. For this purpose, the charge scheduling system 118 is configured to reconcile the predicted driving schedule against the predicted emission data associated with the charging locations 308-312 approached during the predicted journeys 301 a-d. It shall be appreciated that the charge scheduling system 118 may be configured to use one or more schemes, rules, or methods, for performing such objective optimisation. For example, the charge scheduling system 118 may be configured to determine: (i) which of the identified charging locations 308-312 will be within a threshold proximity of the vehicle 100 during each predicted journey 301 a-d, and an estimated time of that intersection; and (ii) the predicted emissions associated with that charging location 308-312 at the estimated time of intersection; such that the charging locations 308-312 can be ranked accordingly based on the indicated emissions In this manner, the charge scheduling system 118 is able to determine the charging location 308-312 associated with the lowest emissions for charging the vehicle 100 during the predicted driving schedule. By way of example, the charge scheduling system 118 may determine that the vehicle 100 will travel within a threshold proximity of: the first charging location 308 during the first predicted journey 301a, the second charging location 310 during the second predicted journey 301b, none of the identified charging locations 304-312 during the third precited journey 301c, and the first and third charging locations 308, 310 during the fourth predicted journey 304a. The charge scheduling system 118 may further compare the predicted emissions data associated with each charging location 308-312 when the vehicle 100 is predicted to travel within the threshold proximity of that charging location 308-312, i.e. comparing the predicted emissions data associated with: (i) a first one of the charging locations 308 at a time of 07:40 to 07:45, (ii) a second one of the charging locations 310 at a time of 08:20 to 08:25, (iii) a third one of the charging locations 312 at a time of 17:45 to 17:50, and (iv) the first charging location 308 at a time of 17:25 to 17:30. The charge scheduling system 118 may therefore rank the respective charging options according to the associated emissions and determine one or more suggested times and locations for charging the vehicle 100 during the forecast period. The determined charging schedule may be output to a user as a suggestion or used to actively control the vehicle 100 or vehicle navigation. For example, the control system 102 may be configured to output one or more control signals to the user interface 120 for communicating the charging schedule, or individual suggestions thereof. Additionally, or alternatively, the control system 102 may be configured to output one or more control signals to the navigation system 124, for example to update the route guidance, based on the determined charging schedule. A method of determining a charging schedule for the vehicle 100, using the control system 102, shall now be discussed with additional reference to Figure 4. Figure 4 shows an exemplary method 400 of determining the charging schedule for the vehicle 100, in accordance with an embodiment of the disclosure. It shall be appreciated that the journey forecasting system 114 has been suitably trained, configured, and / or calibrated for determining predicted driving schedules for the user I vehicle 100 at this stage, and is therefore operating in the active stage. For example, the control system 102 may have already obtained reference journey data in a preceding calibration stage and the behavioural model 114a may be trained or calibrated accordingly for predicting a driving schedule for the user I vehicle 110. In step 402, the journey forecasting system 114 therefore obtains a forecast period for the charging schedule. For example, the journey forecasting system 114 may obtain the forecast period based one or more user inputs, entering a corresponding value (i.e., a length of time or two separate dates, as mentioned above) into the user interface 120. Alternatively, the journey forecasting system 114 may access pre-programmed instructions for setting the forecast period, stored in the memory 122. The forecast period is the time period for which the charging schedule will be determined. For example, if the forecast period is defined as a period of one week, then the charging schedule will be determined to span the course of one week from the day that the system is instructed, including the day that the forecast period is obtained. That is, if the forecast period is obtained on Monday 24 June, then the charging schedule will be determined for the period between Monday 24 June 2024 to Sunday 30 June 2024 inclusive. Alternatively, as mentioned above, the forecast period may be two separate dates which bound a time period such as 1 March 2024 to 10 March 2024, in which case the charging schedule would be determined for between 1 March 2024 to 10 March 2024. As mentioned previously, in some examples the forecast period may be determined and updated iteratively, for example at prescribed frequency. For the purpose of the following description, the forecast period obtained, in step 402, is set to the ‘next day’ to continue the simplified example presented previously. It shall be appreciated that this example is not intended to be limiting on the scope of the invention though. In step 404, the journey forecasting system 114 determines a predicted driving schedule for the forecast period. That is, the journey forecasting system 114 determines the predicted journeys that the vehicle 100 is expected to complete during the forecast period, including the date, time and route information for producing a schedule of journeys expected to be completed by the user or vehicle 100 during the forecast period. As described above, the journey forecasting system 114 is configured to determine the predicted driving schedule using a behavioural model 114a, which has been trained or calibrated to predict vehicle journeys based on reference journey data obtained during the preceding calibration stage. This reference journey data may include identified preferred geographic locations, i.e. locations of significance to the user which may be visited frequently, such as home and work locations, frequently visited shops, cafes, etc. In step 404, the journey forecasting system 114 may also account for a desired or determined charging frequency for the forecast period in order to determine the predicted driving schedule. The forecast period obtained, in step 402, may therefore be provided as an input to the behavioural model 114a and the behavioural model 114a may therefore output a predicted driving schedule, in step 404. To continue the previous example, the behavioural model 114a may therefore output a predicted driving schedule, which includes the first to fourth predicted journeys 301 a-d discussed previously in relation to Figure 3. In step 406, the journey forecasting system 114 determines the predicted journey area 302 based on the plurality of predicted journeys 301 a-d. As noted previously, the predicted journey area 302 is a bounding geographical area that encompasses the predicted journeys 301 a-d and the journey forecasting system 114 may use one or more enveloping or fitting functions for this purpose. As shown in Figure 3, the predicted journey area 302 may therefore be determined as a circular or elliptical area that is sized and positioned so as to encompass the plurality of predicted journeys 301 a-d, for example with a threshold margin extending along or around the routes of the predicted journeys 301a-d. Again, for this purpose, the journey forecasting system 114 may use the geographic co-ordinates of the predicted journeys 301 a-d for the purpose of determining the predicted journey area 302. In step 408, the journey forecasting system 114 determines or identifies one or more charging locations 308-312 located within the predicted journey area 302. The identified charging locations may be publicly available charging stations and the journey forecasting system 114 may access a database of charging stations, containing geographically referenced information for the purpose of identifying those charging stations 308-312. In the present example, the journey forecasting system 114 therefore identifies the first, second, and third charging locations 308-312 shown in Figure 3. In step 410, the emissions forecasting system 116 determines predicted emissions data for each of the charging locations 308-312 identified by the journey forecasting system 114, in step 408. The predicted emissions data may be determined for all, or a relevant portion, of the forecast period. For example, the predicted emissions data may only be determined for respective charging locations 308-312 for periods during which the predicted driving schedule indicates that the vehicle 100 will be within a threshold proximity of those charging locations 308-312. For each of the identified charging locations 308-312, the predicted emission data may be determined by accessing such information provided on a separate server, such as a third-party server, or the predicted emission data may be determined by the emissions forecasting system 116 itself, for example by applying one or more time series forecasting methods. For example, the emissions forecasting system 116 may be configured to receive, or recall from the memory 122, historic emissions data for each of the identified charging locations 308-312 and apply one or more time-series forecasting methods to determine predicted emissions data based on the historic emissions data. The precited emissions data allows the control system 102 to predict relative emission intensities (such as carbon intensity) for the respective charging locations 308-312, particularly when the predicted driving schedule indicates that the vehicle 100 will be within a threshold proximity of such charging locations 308-312. For example, in step 410, the emissions forecasting system 116 may determine predicted emissions data for each of the following: (i) the first charging location 308 at a time of 07:40 to 07:45, (ii) the second charging location 310 at a time of 08:20 to 08:25, (iii) the third charging location 312 at a time of 17:45 to 17:50, and (iv) the first charging location 308 at a time of 17:25 to 17:30. The predicted emissions data may therefore be indicative of the relative emission intensity of the respective charging locations 308-312 as encountered during the predicted driving schedule. In step 412, the charge scheduling system 118 determines the charging schedule based on the predicted driving schedule determined in step 404, the charging locations identified in step 408, and the emissions data predicted in step 410. The charge scheduling system 118 may also determine the charging schedule based on the charging frequency determined in step 404. The charge scheduling system 118 may be configured to minimise the emissions associated with charging the vehicle 100 and use one or more schemes, rules, or methods, for performing such objective optimisation. For example, the charge scheduling system 118 may be configured to compare the predicted emission data associated with each charging location 308-312 when the vehicle 100 is predicted to travel within a threshold proximity of that charging location 308-312. The charge scheduling system 118 may therefore determine the charging schedule based on the relative emission intensity indicated for the respective charging locations 308-312. Continuing the previous example, the charge scheduling system 118 may therefore determine that the minimum emission intensity is produced when the first predicted journey 301a passes the first charging location 308, at a time of 07:40 to 07:45, and determine a corresponding charging schedule with a suggestion to charge the vehicle 100 at the first charging location 308 during the first predicted journey 301a. If the charging frequency is a factor in determining the charging schedule, then the charge scheduling system 118 may be configured to include recommended charge time(s) and corresponding location(s) in the charging schedule, which correlate to the charging frequency. In step 414, the control system 102 may therefore output one or more control signals based on the determined charging schedule. For example, the control system 102 may output one or more control signals to the user interface 120 to control the user interface 120 to display the determined charging schedule, or individual suggestions thereof. For example, the control system 102 may output one or more control signals to the user interface 120 to display a suggestion to charge the vehicle 100 at the first charging location 308 at 07:40. The suggestion may be output on the display in advance of the forecast period, or during the forecast period, for example whilst the vehicle 100 is completing the first predicted journey 301a. The suggestions informs the user of ideal charging times I location and allows the user to view and choose individual routes to take equipped with such information. In other examples, the control system 102 may output one or more control signals to the navigations system 124 instead, for example to update the route guidance automatically, or as a suggested deviation, to direct the vehicle 100 to the suggested charging location. In this manner, the suggested times and charging locations may be optimised for reduced carbon emissions, whilst being complementary to the expected journeys that the user / vehicle would follow as a habit of their ordinary driving behaviours. It is envisaged that the control system 102 will therefore lead to enhanced utilisation of low-emission energy sources for charging the vehicle 100, ultimately reducing the carbon footprint associated with the vehicle 100 over its operating lifecycle. In each case, it shall be appreciated that the method 400, or steps thereof, may be performed iteratively, for example at a predetermined frequency, upon receiving a new forecast period or upon determining changes to the driving behaviour or predicted driving schedule. It shall be appreciated that various changes and modifications can be made to the examples described above without departing from the scope of the present invention. In particular, it shall be appreciated that the charge scheduling system 118 may be configured to take one or more additional factors into account in determining the charging schedule in other examples. For example, the control system 102 may be configured to determine the charging schedule irrespective of a current state of charge of the vehicle 100, substantially as described above. However, in other examples, the control system 102 may be configured to account for the effect of the current state of charge on the driving behaviour. The charge scheduling system 118 may therefore be further configured to receive signals indicative of a state of charge or range of the vehicle 100, and determine a predicted state of charge or range of the vehicle 100 when following the predicted journeys 301a-d. In which case, it shall be appreciated that the determined charging schedule may be adapted based on the predicted state of charge of the vehicle 100, for example generating a less preferred charging suggestion (i.e. at a less preferred location or time) when charging is needed to maintain a state of charge of the vehicle 100 that is greater than or equal to a lower threshold state of charge or minimum charge. The minimum charge may therefore be used as part of a multi-objective optimisation analysis performed by the charge scheduling system 118 to determine the charging schedule. In the examples described above, it shall be appreciated that the control system 102 is configured to identify public charging stations for charging the vehicle 100. However, it shall be appreciated that the invention it not limited in this respect and, in other examples, the control system 102 may be configured to further consider private charging locations, 5 such as at the home location 304 or work location 306 of the user. In this respect, the control system 102 operates substantially as described previously, further determining predicted emissions data for the private charging locations and updating the charging schedule accordingly if it is determined that one of the private charging locations is associated with the lowest emissions for charging. 10

Claims

1. A computer implemented method of determining a charging schedule for a battery of an electric drive system of a vehicle, the method comprising:obtaining a forecast period for a charging schedule;determining a predicted driving schedule comprising a plurality of predicted journeys for the forecast period using a behavioural model configured to model driving behaviour of the vehicle;determining a predicted journey area based on the plurality of predicted journeys;identifying one or more charging locations located within the predicted journey area;determining predicted emissions data associated with the one or more charging locations for the forecast period; anddetermining the charging schedule based on:the predicted driving schedule;the identified one or more charging locations; andthe predicted emissions data associated with the one or more charging locationswherein the determined charging schedule comprises one or more suggested times and locations for charging the vehicle during the forecast period.

2. A method according to claim 1, wherein the behavioural model is a statistical model calibrated based on reference journey data indicative of the driving behaviour of the vehicle.

3. A method according to claim 1, wherein the behavioural model is a machine learning model trained based on reference journey data indicative of the driving behaviour of the vehicle.

4. A method according to claim 2 or claim 3, wherein the reference journey data comprises historic journey data indicative of a plurality of historic journeys completed by the vehicle.

5. A method according to claim 4, wherein the historic journey data is indicative of the date, time, and routes of each historic journey.

6. A method according to claim 5, wherein the historic journey data comprises one or more of the following for each historic journey:a start location and an end location;a start time and an end time;a route between the start and end locations; and / ora state of charge (SoC) of the vehicle and a predicted remaining range of the vehicle at the start and end location.

7. A method according to any of claims 4 to 6, when dependent on claim 3, wherein the method further comprises the steps of: obtaining the historic journey data during a calibration stage; andtraining the behavioural model based on the historic journey data prior to determining the predicted driving schedule.

8. A method according to any of claims 2 to 7, wherein the method further comprises the step of receiving reference journey data via a user input device of the vehicle.

9. A method according to any preceding claim, wherein determining the predicted emissions data comprises applying one or more time series forecasting methods to historic emissions data for each of the identified charging locations, wherein the historic emission data comprises time series data of the power source and the associated emissions.

10. A method according to any preceding claim, wherein the method further comprises outputting the charging schedule to a user of the vehicle.

11. A method according to any preceding claim, wherein determining the charging schedule comprises:determining which of the identified charging locations is located within a threshold proximity of the vehicle according to each predicted journey;determining an estimated time of intersection of the vehicle with that charging location; anddetermined the predicted emissions associated with that charging location at the estimated time of intersection.

12. A method according to any preceding claim, wherein the obtained forecast period is updated at a prescribed frequency such that the charging schedule is redetermined for the updated forecast period.

13. A method according to any preceding claim, further comprising the step of obtaining data indicative of a range of the vehicle; wherein the charging schedule is determined based, at least in part, on the indicated range of the vehicle.

14. An energy management control system for a vehicle comprising an electric drive system and a battery for the electric drive system, the control system being configured to determine a charging schedule for the battery, the control system comprising one or more controllers configured to:obtain a forecast period for the charging schedule;determine a predicted driving schedule comprising a plurality of predicted journeys for the forecast period using a behavioural model configured to model driving behaviour of the vehicle;determine a predicted journey area based on the plurality of predicted journeys;identify one or more charging locations located within the predicted journey area;determine predicted emissions data associated with the one or more charging locations for the forecast period; anddetermine the charging schedule based on:the predicted driving schedule;the identified one or more charging locations; andthe predicted emissions data associated with the one or more charging locations;wherein the determined charging schedule comprises one or more suggested times and locations for charging the vehicle during the forecast period.

15. A vehicle comprising an energy management control system according to claim 14.

Images