Description:Complete Specification:
The following specification describes and ascertains the nature of this invention and the manner in which it is to be performed:


Field of the invention:
The present disclosure relates a device to estimate a load of the vehicle and a method thereof.

Background of the invention:
In Commercial vehicle segment, vehicles run to-and-fro from one point (pick-up point) to another (drop-off point). In almost all cases one of the trip will be either fully loaded or fully unloaded. The fleet owner will not have correct information whether vehicle is loaded or unloaded. The owner has to rely on information given by the driver. In case if this information is not true, the owner has to bear the excess cost (like cost of diesel / maintenance / demurrage / driver etc.).

According to a prior art WO13075280 a vehicle mass estimation method and system is disclosed. In a vehicle mass estimation method and system, wheel speed, driving torque and longitudinal acceleration of the vehicle are obtained, and an estimated vehicle mass is calculated using an estimation equation group which comprises wheel speed, driving torque and longitudinal acceleration as input parameters, and vehicle mass and driving resistance of the vehicle as variables. Lower and upper thresholds are considered in the calculating process of the estimated vehicle mass.

Brief description of the accompanying drawings:
An embodiment of the disclosure is described with reference to the following accompanying drawings,
Fig. 1 illustrates a block diagram of a device to estimate load of a vehicle, according to an embodiment of the present invention, and
Fig. 2 illustrates a flow diagram of a method for estimating load of the vehicle, according to the present invention.

Detailed description of the embodiments:
Fig. 1 illustrates a block diagram of a device to estimate load of a vehicle, according to an embodiment of the present invention. The device 100 comprises a controller 110 configured to receive real-time input data 128 comprising primary parameters of the vehicle 130 measured by the in-vehicle sensors and secondary parameters comprising a rate of change engine speed, a current gear position and engine torque. The secondary parameters are instantaneously derived/computed using the measured primary parameters. The primary parameters comprises an engine speed, a vehicle speed, an accelerator pedal position and an inclination of the vehicle 130 measured by respective sensors such as an engine speed sensor 102, a wheel speed sensor 104, a position sensor 106 and inclination sensor 108. The device 100, characterized in that, the controller 110 further configured to check the fulfillment of preset boundary conditions 116 for a predetermined set of parameters from the real-time input data 128. The controller 110 then processes the real-time input data 128 through a load estimation model 118 upon satisfaction with the preset boundary conditions 116 and estimate/predict load of the vehicle 130.

According to an embodiment of the present invention, the load estimation model 118 is a physics based model comprising an equation which considers force at the wheel is equal to force due to inertia, force due to rolling resistance, force due to air drag and force due to inclination.
F_wheel=F_inertia+F_(Rolling resistance)+F_(Air drag)+F_Inclination
where,
F_wheel=(Wheel torque)/(Tyre radius)
F_inertia=M*a
F_(Rolling resistance)=C_r*M*g*cos??
F_(Air drag)=0.5*C_d*A*?*V^2
F_Inclination=M*g*sin??
M=(F_wheel-0.5*C_d*A*?*V^2)/((a+C_r*g*cos?? )+g*sin?? )
where,
M= Mass of the vehicle (Kg)
a = acceleration (m/s2)
g = Acceleration due to gravity (m/s2)
? = Inclination angle (rad)
Cr = Coefficient of rolling resistance
Cd = Coefficient of air resistance
A = Vehicle frontal area (m2)
? = Air density (Kg/m2)
V = Vehicle velocity (m/s)

The controller 110 uses the equation of M in estimating the load of the vehicle 130 in real time and the same is classified as full load, partial load and no load and provided as output 114. In other words, the estimated load is classified as any one of a no load, partial load and full load by the controller 110.

According to an embodiment of the present invention, the gear position is determined using a trained machine learning algorithm which uses the ratio of engine speed and the vehicle speed measured by engine speed sensor 102 and the wheel speed sensor 104 respectively.

According to an embodiment of the present invention, the boundary conditions 116 comprise comparison of the rate of change of engine speed to be greater than a threshold engine speed, the gear position to be greater than predetermined gear position, the accelerator pedal position to be greater than a threshold position and elevation to be lesser than threshold gradient/elevation. The boundary conditions 116 are checked by the controller 110 of the device 100.

According to embodiment of the present invention, the device 100 is at least once selected from a group comprising an internal device and an external device. The internal device comprises an Engine Control Unit (ECU) 132, a Vehicle Control Unit (VCU), a Connectivity Control Unit (CCU), a Telematic Control Unit (TCU) 120 and the like. The external device comprises the portable computing unit 126 and a cloud computer 124. The portable computing unit 126 is a smartphone, a wearable device, or a laptop and the like. In another alternative, the device 100 is combination of internal device and external device in manner that the processing is shared based on load of ongoing tasks.

In accordance to an embodiment of the present invention, the controller 110 is provided with necessary signal detection, acquisition, and processing circuits. The controller 110 is the control unit which comprises input/output interfaces having pins or ports, the memory element 112 such as Random Access Memory (RAM) and/or Read Only Memory (ROM), Analog-to-Digital Converter (ADC) and a Digital-to-Analog Convertor (DAC), clocks, timers, counters and at least one processor (capable of implementing machine learning) connected with each other and to other components through communication bus channels. The memory element 112 is pre-stored with logics or instructions or programs or applications or modules/models and/or threshold values/ranges, system threshold, predefined/predetermined criteria/conditions, engine maps/table which is/are accessed by at least one processor as per the defined routines. The internal components of the controller 110 are not explained for being state of the art, and the same must not be understood in a limiting manner. The controller 110 may also comprise communication units to or may be connected to other control units to communicate through wireless or wired means such as Global System for Mobile Communications (GSM), 3G, 4G, 5G, Wi-Fi, Bluetooth, Ethernet, serial networks, and the like. The controller 110 is implementable in the form of System-in-Package (SiP) or System-on-Chip (SOC) or any other known types. Examples of controller 110 comprises but not limited to, microcontroller, microprocessor, microcomputer, etc.

According to an embodiment of the present invention, the device 100 is a cloud computer 124. The cloud computer 124 comprises the controller 110 and is part of the cloud computer 124 based solution. The cloud computer 124 receives the data which is transmitted by the TCU 120 of the vehicle 130 through mobile/cellular network 122. The TCU 120 receives the real time input data 128 through the communication network of the vehicle 130. Alternatively, the real-time data 128 is collected by a smartphone or portable computing unit 126 which is in communication with the vehicle 130 through wireless or wired means known in the art, and the smartphone in-turn transmits the collected data to the cloud computer 124 through the mobile network 122. The cloud computer 124 then performs the necessary operation as already defined and estimates the load of the vehicle 130.

According to an embodiment of the present invention, the device 100 is the ECU 132 (or VCU) of the vehicle 130. The ECU 132 comprises the controller 110 which receives the real-time input parameters 128 through the in-vehicle sensors through internal communication networks. The controller 110 checks the predetermined parameters from the real-time input data 128 with respect to the boundary conditions 116. Once the boundary conditions 116 are satisfied, the controller 110 processes the real-time input data 128 through the load estimation model 118 and estimates/predicts the load of the vehicle 130. The estimated load is then intimated to the owner of the vehicle 130 through TCU 120.

According to an embodiment of the present invention, the communication network is at least one selected from a group comprising a Controller Area Network (CAN), a Local Interconnect Network (LIN), other vehicular network, Wi-Fi, Bluetooth, or Subscriber Identity Module (SIM) or e-SIM based network through the Telematics Control Unit (TCU) 120 or Connectivity Control Unit (CCU).

Consider a fleet service provider contains fleet of vehicles 130 for taxis, transport, etc. Each of the vehicle 130 is fit with the TCU 120 or other connectivity means. The data from all the vehicles 130, whenever driven, are collected, and stored in the cloud computer 124. The controller 110 in the cloud computer 124 then calculates the secondary parameters. The real-time input data 128 is then checked against the boundary conditions 116. Once the check is fulfilled, the controller 110 processes the real-time input data 128 is passed through the load estimation model 118 to estimate load of the vehicle 130. Further, the controller 110 is able to classify the load as full load, partial load, and no load.

According to the present invention, the vehicle 130 comprises but not limited to two-wheeler, three wheeler vehicles, four wheeler vehicles, Off-Highway (OHW) vehicle, lorries, load carrying vehicles, commercial vehicles.

According to the present invention, a working of the device 100 is explained. The device 100 is considered to be the cloud computer 124. The raw data or real-time input data 128 from TCU 120 is received by the controller 110 in the cloud computer 124. The raw data is pre-processed in the cloud computer 124, by the controller 110 to derive or compute the secondary parameters. The controller 110 checks for the boundary conditions 116 to ensure vehicle 130 is not running in over run condition and not running in gradient condition. The real-time input data 128 extracted after boundary conditions 116 are further filtered to remove noisy measurements, and then sent to load estimation model 118 for actual vehicle load predictions. The result of predicted output 114 is then classified under “full load”, “no load” or “part load” based on the determined load.

The primary parameters of the real-time input data 128 from the vehicle 130 from the TCU 120 is pushed to the controller 110 where, in a pre-processing layer, the secondary parameters are computed/extracted. The rate of change of engine speed is used to know if vehicle 130 is in accelerating or decelerating, because only accelerating condition data is considered as deceleration will not represent the vehicle load conditions. The gear position is not available from the vehicle OBD data. Hence, a Machine Learning (ML) algorithm is used to cluster the vehicle speed to engine speed ratio and predict the current gear position of the vehicle 130. Here, the vehicle 130 must run in all gears until sufficient data is available to predict all gears. The gear position is used to consider the input data 128 for the vehicle running in second gear and above, because the first gear data is not considered as there will be instances of half clutch.

The real-time input data 128 comprising the primary parameters and secondary parameters, is pushed to boundary condition layer of the controller 110. An example of the boundary conditions 116 are provided and the same must not be understood in limiting sense. The rate of change of engine speed > 0 rpm/s, the gear position >1, the accelerator pedal position > 1% (to ensure vehicle in not in overrun condition), the elevation < 30 to ensure vehicle is not running in gradient road conditions. After fulfilling boundary conditions 116, the real-time input data 128 is pushed to filter and load estimation model 118. Already known filtering techniques are used to remove noisy measurements and estimate vehicle parameter actual engine torque close to accurate value. All the real-time input parameters 128 are then pushed to the load estimation model 118. The load estimation model 118 (load prediction model) is physics-based model where individual forces acting on vehicle 130 are calculated.

The air drag and rolling resistance forces are constant for a given make and model of the vehicle 130. The effect of inclination force is removed as data operating in flat road conditions are considered, not gradient road conditions. The estimated/predicted results are averaged out to stabilize the output. This stabilized output 114 is classified as “full load”, “no load” or “part load” accordingly.

Fig. 2 illustrates a flow diagram of a method for estimating load of the vehicle, according to the present invention. The method comprises plurality of steps, of which a step 202 comprises receiving, by the controller 110, real-time input data 128 comprising primary parameters of the vehicle 130 measured by the in-vehicle sensors and secondary parameters comprising the rate of change engine speed, the current gear position and engine torque. The method is characterized by a step 204 which comprises checking, by the controller 110, the fulfillment of preset boundary conditions 116 for the predetermined set of parameters from the real-time input data 128. A step 206 comprises processing, by the controller 110, the real-time input data 128 through the load estimation model 118 upon satisfying the preset boundary conditions 116 and estimating load of the vehicle 130. The method is executed or performed by the controller 110.

According to the method, the gear position is determined using the trained Machine Learning (ML) model. The boundary conditions 116 comprises the rate of change of engine speed greater than threshold engine speed, the gear position is greater than pre-determined gear position, the accelerator pedal position is greater than the threshold position, and elevation is lesser than the threshold gradient. The load estimation model 118 is the physics based model.

According to the present invention, the method comprises classifying the estimated load as any one of a no load, partial load and full load.

According to the method, the device 100 is at least one selected from the group comprising the internal device and the external device. The internal device comprises the Engine Control Unit (ECU) 132, the Vehicle Control Unit (VCU), the Connectivity Control Unit (CCU), the Telematics Control Unit (TCU) 120. The external device comprises the portable computing unit 126 and the cloud computer 124.

According to the present invention, real time vehicle load detection using only On-Board Diagnostics (OBD) parameters and without adding any sensors. The present invention focuses on providing real-time status without addition of any sensors, where it predicts vehicle load and classify as “full load”, “no load” or “part load”.

It should be understood that embodiments explained in the description above are only illustrative and do not limit the scope of this invention. Many such embodiments and other modifications and changes in the embodiment explained in the description are envisaged. The scope of the invention is only limited by the scope of the claims.
, Claims:We claim:
1. A device (100) to estimate load of a vehicle (130), said device (100) comprises a controller (110) configured to,
receive real-time input data (128) comprising primary parameters of said vehicle (130) measured by in-vehicle sensors and secondary parameters comprising a rate of change engine speed, a current gear position and an engine torque, characterized in that,
check fulfillment of preset boundary conditions (116) for a predetermined set of parameters from said real-time input data (128), and
process said real-time input data (128) through a load estimation model (118) upon satisfaction with said preset boundary conditions (116) and estimate load of said vehicle (130) as output (114).

2. The device (100) as claimed in claim 1, wherein said gear position is determined using a trained Machine Learning (ML) model, and wherein said load estimation model (118) is a physics based model.

3. The device (100) as claimed in claim 1, wherein said boundary conditions (116) comprises said rate of change of engine speed greater than threshold engine speed, said gear position greater than pre-determined gear position, said accelerator pedal position greater than a threshold position, and elevation is lesser than threshold angle.

4. The device (100) as claimed in claim 1, wherein said estimated load is classified as any one of a no load, partial load and full load.

5. The device (100) as claimed in claim 1 is at least one selected from a group comprising an internal device and an external device, wherein said internal device comprises an Engine Control Unit (ECU) (132), a Vehicle Control Unit (VCU), a Connectivity Control Unit (CCU), a Telematics Control Unit (TCU) (120), and said external device comprises a portable computing unit (126) and a cloud computer (124).

6. A method for determining load of a vehicle (130), said method comprising the steps of:
receiving real-time input data (128) comprising primary parameters of said vehicle (130) measured by in-vehicle sensors and secondary parameters comprising a rate of change engine speed, a current gear position and an engine torque, characterized by,
checking fulfillment of preset boundary conditions (116) for a predetermined set of parameters from said real-time input data (128), and
processing said real-time input data (128) through a load estimation model (118) upon satisfying said preset boundary conditions (116) and determining load of said vehicle (130) as output (114).

7. The method as claimed in claim 6, wherein said gear position is determined using a trained Machine Learning (ML) model, and said load estimation model (118) is a physics based model.

8. The method as claimed in claim 6, wherein said boundary conditions (116) comprises said rate of change of engine speed greater than threshold engine speed, said gear position is greater than pre-determined gear position, said accelerator pedal position is greater than a threshold position, and elevation is lesser than threshold gradient.

9. The method as claimed in claim 6, comprises classifying said estimated load as any one of a no load, partial load and full load.

10. The method as claimed in claim 6 is at least one selected from a group comprising an internal device and an external device, wherein said internal device comprises an Engine Control Unit (ECU) (132), a Vehicle Control Unit (VCU), a Connectivity Control Unit (CCU), a Telematics Control Unit (TCU) (120), and said external device comprises a portable computing unit (126) and a cloud computer (124).