VEHICLE

Abstract

The present invention relates to a vehicle (100) with a detection unit (103) for detecting trend data of an operating state of the vehicle (100); and a prediction unit (105) for predicting a vehicle parameter based on a pattern comparison between the detected trend data and predetermined comparison data.

Description

[0001] The invention relates to a vehicle and a method for operating a vehicle.

[0002] Prediction algorithms for optimizing a specific system require precise predictions of the future load profile. In the automotive sector, for example, it may be important to predict a changing vehicle mass in order to determine an optimal power distribution in hybrid vehicles (energy optimization). In the application case of the operational optimization of a thermal system, it is advantageous to predict a thermal load in order to implement heating or cooling in a predictively efficient manner. If the amount of energy required for heating or cooling is known, a hybrid system can be operated accordingly, and a battery state of charge can be predictively and efficiently specified. This applies to all predictive optimization problems that extend beyond a short time horizon.

[0003] The vehicle according to the invention comprises a detection unit for detecting historical data of an operating state of the vehicle; and a prediction unit for predicting a vehicle parameter based on a pattern comparison between the detected historical data and predefined comparison data. This enables an increase in the prediction accuracy of vehicle parameters and an improvement in the optimality of predictive algorithms.

[0004] In an advantageous embodiment of the vehicle, the operating state is a vehicle mass, a thermal output, a cooling output, a temperature, and / or an electrical output. This allows particularly suitable operating states to be used for comparison.

[0005] In a further advantageous embodiment of the vehicle, the vehicle parameter is a vehicle mass, a heating output, and / or a cooling output. However, the vehicle parameter can also be the operating parameter.

[0006] This makes it possible to determine the parameters mentioned precisely and use them for further control of the vehicle.

[0007] In a further advantageous embodiment of the vehicle, the historical data comprises time data and / or position data of the vehicle. The time data comprises, for example, consecutive times, and the position data comprises consecutive positions, each of which is assigned to the operating states. This enables the accuracy of the prediction to be further improved.

[0008] In a further advantageous embodiment of the vehicle, the vehicle comprises a memory for storing the comparison data. This enables the comparison data to be quickly available.

[0009] In a further advantageous embodiment of the vehicle, the vehicle includes a wireless interface for retrieving the comparison data. This enables the retrieval of current comparison data.

[0010] In a further advantageous embodiment of the vehicle, the prediction unit comprises a neural network for predicting the vehicle parameter. This enables pattern recognition to be performed easily.

[0011] In a further advantageous embodiment of the vehicle, the prediction unit comprises a processor. This enables the prediction unit to process the data.

[0012] In a further advantageous embodiment of the vehicle, the detection unit comprises a sensor for detecting the operating state. This enables the operating state to be reliably detected.

[0013] The method according to the invention for operating a vehicle comprises the steps of acquiring historical data of an operating state of the vehicle; and predicting a vehicle parameter based on a pattern comparison between the acquired historical data and predefined comparison data. Subsequently, in a further step, the predicted vehicle parameter can be used in a closed-loop or closed-loop control of the drive system or auxiliary consumers, such as, for example, in a heating system.

[0014] Embodiments of the invention are explained in more detail with reference to the following figures. Fig. 1 is a schematic representation of a vehicle with a prediction unit; and Fig. 2 a block diagram of a method for operating a vehicle.

[0015] Fig. 1 shows a schematic representation of a vehicle 100 with a detection unit 103 for detecting historical data of an operating state of the vehicle 100 and a prediction unit 105 for predicting a vehicle parameter based on a pattern comparison between the detected historical data and predetermined comparison data.

[0016] The vehicle 100 is, for example, an automobile, a truck, a delivery van, a bus, a train, a construction vehicle, or a commercial vehicle. For this purpose, the prediction unit 105 comprises a digital memory 107 for storing data and programs and a processor 111 that can process the respective data and execute suitable pattern matching algorithms.

[0017] For pattern recognition, suitable algorithms can be used that can recognize recurring patterns in the operating state history data. For example, trained neural networks can be used to map the operating state history to the vehicle parameter to be predicted. Training data sets are used for this purpose, which include not only the operating state history but also the respective vehicle parameter. After training, the neural network is able to predict the vehicle parameter based solely on the operating state history and / or on environmental variables (supervised learning).

[0018] The detection unit 103 comprises, for example, a sensor 113 for detecting relevant vehicle environmental or operating states of the vehicle 100, such as a vehicle mass, a heat or cooling output (for example, by detecting temperatures and load states of the cooling system), an electrical output of the auxiliary consumers (for example, by measuring current), or an outside temperature. In general, however, any sensors that detect any operating state of the vehicle 100 can be used as sensor 113. Models can also be used for this purpose. The vehicle mass is usually estimated or modeled rather than measured in real time in the vehicle.

[0019] The historical data of the operating states is captured, collected, and stored in a digital memory 107 of the vehicle 100. Additionally, this captured historical data may include data about route information, a route position, a calendar date, a day of the week, a time of day, and / or an operating time.

[0020] The acquired historical data of the operating states are compared with stored and predefined comparison data by the prediction unit 105. The comparison data can, for example, be pre-stored in the memory 107 or retrieved via a wireless interface 109, such as a radio network. The comparison data includes data corresponding to the acquired historical data. Furthermore, the comparison data can include data about the vehicle parameters depending on the operating states. The vehicle parameters are linked to the operating states in the comparison data.

[0021] For example, pattern matching can be used to identify recurring events in the historical data of the operating state, thus predicting the desired vehicle parameter. By comparing the historical data for the respective operating parameter with the specified comparison data, the prediction unit 105 can determine future vehicle parameters based on similar patterns.

[0022] For example, if a match or similarity is detected between the recorded historical operating state data and the specified comparison data, the vehicle parameter that follows the identified pattern in the comparison data is used as the most probable value for the vehicle parameter to be predicted. This increases the accuracy of predictions and thus improves the optimality of predictive algorithms in the vehicle sector.

[0023] For example, a similar mass profile of the vehicle 100 along a route may be captured at given times, such as a school bus or a delivery vehicle.

[0024] In addition, heating or cooling requirements can be determined along a route or at specific times of the year, for example, for commuters on their way to work. Special events, such as major events or holidays, can be excluded or explicitly taken into account. The recorded data can be compared with environmental or surrounding data, such as a bus route on holidays or during major events. For example, a holiday may have the same mass pattern for a bus route as a Sunday and not as a regular weekday on the same day of the week. From these past patterns, the vehicle parameters can be predicted if similar patterns occur.

[0025] Fig.Figure 2 shows a block diagram of the method for operating vehicle 100. In step S101, the historical data of the operating state of vehicle 100 is acquired. In step S102, the vehicle parameter is calculated and / or predicted based on a pattern comparison between the acquired historical data and predefined comparison data. Comparison data can be determined using pattern recognition, and a pattern can subsequently be identified to make a further prediction. The operating state itself can also be a vehicle parameter to be predicted.

[0026] All features explained and shown in connection with individual embodiments of the invention can be provided in different combinations in the subject matter according to the invention in order to simultaneously realize their advantageous effects.

[0027] All method steps can be implemented by devices suitable for performing the respective method step. All functions performed by physical features can be a method step of a method.

Claims

[1] Vehicle (100), with; - a recording unit (103) for recording historical data of an operating state of the vehicle (100); and - a prediction unit (105) for predicting a vehicle parameter based on a pattern comparison between the acquired historical data and predetermined comparison data. [2] The vehicle (100) of claim 1, wherein the operating state is a vehicle mass, a heat output, a cooling output, a temperature, and / or an electrical power. [3] Vehicle (100) according to one of the preceding claims, wherein the vehicle parameter is a vehicle mass, a heating power and / or a cooling power. [4] Vehicle (100) according to one of the preceding claims, wherein the historical data comprises time data and / or position data of the vehicle (100). [5] Vehicle (100) according to one of the preceding claims, wherein the vehicle (100) comprises a memory (107) for storing the comparison data. [6] Vehicle (100) according to one of the preceding claims, wherein the vehicle (100) comprises a wireless interface (109) for retrieving the comparison data. [7] Vehicle (100) according to one of the preceding claims, wherein the prediction unit (105) comprises a neural network for predicting the vehicle parameter. [8] Vehicle (100) according to one of the preceding claims, wherein the prediction unit (105) comprises a processor (111). [9] Vehicle (100) according to one of the preceding claims, wherein the detection unit (105) comprises a sensor (113) for detecting the operating state. [10] A method for operating a vehicle (100), comprising the steps of: - recording (S101) historical data of an operating state of the vehicle (100); and - Predicting (S102) a vehicle parameter based on a pattern comparison between the recorded historical data and specified comparison data.

Images