An energy prediction-based unmanned aerial vehicle precise recovery method and system

Abstract

This application discloses a method and system for precise drone recovery based on energy prediction. The method specifically includes the following steps: setting a flight path; predicting the energy at the parachute deployment point based on the set flight path; issuing a parachute deployment command based on the predicted energy at the deployment point; and completing the recovery by reaching a preset location based on the deployment command. The drone recovery method proposed in this application can effectively improve the accuracy of the drone's parachute deployment point position and speed. Compared with other recovery strategies in the prior art, the drone recovery method proposed in this application improves control accuracy and recovery speed.

Description

Technical Field

[0001] This application relates to the field of unmanned aerial vehicles (UAVs), specifically to a method and system for precise recovery of UAVs based on energy prediction. Background Technology

[0002] With my country's increasing population and decreasing available unmanned areas, the space available for drone recovery is shrinking. Meanwhile, the explosive growth of the drone market and the increasing number of launches and recoveries place higher demands on recovery speed and parachute deployment accuracy. Precise recovery strategies can effectively reduce the dispersion of drone landing sites, decrease the requirements for recovery areas, and save search time after landing. Currently, drones using parachute + airbag recovery systems employ two recovery strategies: Strategy 1: Fixed-point parachute deployment: After reaching a designated waypoint Xn, the drone shuts down its engine and deploys its parachute; Strategy 2: Fixed-point shutdown, constant-speed parachute deployment: After reaching the designated waypoint Xn, the drone shuts down its engine and glides without power until its speed reaches a set deployment speed Vn, at which point it deploys its parachute. These two strategies, one only constraining the deployment point and not the recovery speed, and the other only constraining the deployment speed and not the deployment point, both fail to achieve precise recovery. Specifically, Strategy 1: Fixed-point parachute deployment means the drone shuts down its engine and deploys its parachute after reaching the designated waypoint Xn. The main drawback is the uncontrollable parachute deployment speed. This strategy only controls the deployment speed, but does not constrain it. The deployment speed mainly depends on the speed Vn-1 at waypoint Xn-1 and the distance R between waypoints Xn-1 and Xn. If the speed Vn-1 is too high and the distance R is too small, the UAV will not have enough distance to decelerate, resulting in an excessively high deployment speed. If the deployment speed is too high, it will cause significant impact overload on the UAV's structure and related equipment, and may even lead to the risk of disintegration. Strategy Two, fixed-point shutdown and constant-speed parachute deployment, involves the UAV flying to the set waypoint Xn (shutdown speed Vn), shutting down the engine, and gliding without power to the deployment speed V0 before deploying the parachute. The main drawbacks are: the parachute deployment point is uncontrollable. This strategy only constrains the parachute deployment speed, and the deployment point location mainly depends on the shutdown speed Vn. If Vn is large, the drone will glide without power for a long distance, resulting in a large distance between the deployment point and the waypoint Xn. This longer glide distance requires a larger recovery range area. These two recovery strategies, by constraining only one of the recovery speed or the recovery point (parachute deployment point), cannot achieve precise drone recovery.

[0003] Therefore, how to provide a method for recovering drones with improved accuracy is a problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0004] This application proposes a method for recovering unmanned aerial vehicles (UAVs) based on energy prediction, which includes the following steps: setting a flight path; predicting the energy at the parachute deployment point based on the set flight path; issuing a parachute deployment command based on the predicted parachute deployment energy; and completing the recovery by reaching a preset location based on the parachute deployment command.

[0005] As shown above, the route includes multiple waypoints and parachute deployment points.

[0006] As shown above, the prediction of parachute deployment point energy based on the set flight path includes the following sub-steps: determining whether a specified state has been reached; if the UAV reaches the specified state, then predicting the parachute deployment point energy; if the UAV does not reach the specified state, then continuing flight.

[0007] As mentioned above, determining whether a specified state has been reached includes determining whether the drone has reached a specified location or whether the drone's flight speed has reached a specified flight speed.

[0008] As mentioned above, predicting the parachute deployment point energy based on the set flight path includes obtaining the set parachute deployment point energy.

[0009] As shown above, the energy at the parachute deployment point is specifically represented as follows:

[0010]

[0011] Where V0 is the set parachute deployment speed, and m represents the mass of the drone.

[0012] A drone recovery system based on energy prediction specifically includes: a setting unit and a command issuing unit; the setting unit is used to set the flight path; the prediction unit is used to predict the energy at the parachute deployment point based on the set flight path; the command issuing unit is used to issue a parachute deployment command based on the predicted parachute deployment energy, and complete the recovery by reaching a preset position according to the parachute deployment command.

[0013] As described above, the prediction unit specifically includes a judgment module and a prediction module. The judgment module is used to determine whether the UAV has reached the specified state. If the UAV has reached the specified state, the prediction module predicts the energy at the parachute deployment point. If the UAV has not reached the specified state, it continues to fly.

[0014] As shown above, the prediction module includes a parachute deployment point energy acquisition module, which acquires the set parachute deployment point energy.

[0015] An unmanned aerial vehicle (UAV) specifically includes a UAV body and an energy prediction-based UAV recovery system as described in any of the preceding claims.

[0016] This application has the following beneficial effects:

[0017] The UAV recovery method proposed in this application can effectively improve the accuracy of the UAV's parachute deployment point position and speed. Compared with other recovery strategies in the prior art, the UAV recovery method proposed in this application achieves a recovery position control accuracy better than 30m and a recovery speed control accuracy better than 0.1m / s, with significant improvements in both control accuracy and recovery speed. The overall effect is far superior to existing recovery strategies. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this application. For those skilled in the art, other drawings can be obtained based on these drawings.

[0019] Figure 1 This is a flowchart of an energy prediction-based drone recovery method provided according to an embodiment of this application;

[0020] Figure 2 This is a flight path diagram used during the flight of an unmanned aerial vehicle according to an embodiment of this application;

[0021] Figure 3 This is an internal structure diagram of an unmanned aerial vehicle (UAV) recovery system based on energy prediction, provided according to an embodiment of this application. Detailed Implementation

[0022] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0023] With increasingly smaller recovery areas and higher requirements for UAV recovery accuracy, improving recovery accuracy primarily involves enhancing parachute deployment speed and the parachute deployment point. To address this, this application proposes a recovery strategy based on energy prediction. After the Xn-1 waypoint, the UAV predicts the parachute deployment energy based on preset deployment speed and deployment point constraints. When the predicted recovery energy drops below the constraint value, the UAV engine shuts down and taxis to the deployment speed to deploy the parachute. Predicting the recovery energy comprehensively improves the accuracy of both the recovery point and speed.

[0024] Example 1

[0025] like Figure 1 The image shows a method for recovering a drone based on energy prediction, provided in an embodiment of this application. The method specifically includes the following steps:

[0026] Step S110: Set the flight route.

[0027] The flight path includes waypoint Xn-1, which serves as the starting point for energy prediction, and parachute deployment point Xn, such as... Figure 2 As shown, the waypoints in the route include the starting point, waypoint Xn-1 and Xn. The UAV starts from the starting point, passes through waypoint Xn-1 to start energy prediction, and reaches waypoint Xn to open the parachute and complete the UAV recovery.

[0028] Step S120: Predict the energy at the parachute deployment point based on the set flight path.

[0029] The drone flies along a set route, and during the flight, it predicts the drone's parachute deployment point. This includes the following sub-steps:

[0030] Step S1201: Determine whether the specified state has been reached.

[0031] The determination of whether a specified state has been reached includes determining whether the drone has reached a specified location or a specified speed. If the drone has reached a specified location or a specified speed, then step S1202 is executed.

[0032] For example, if the drone flies to waypoint Xn-1, then step S1202 can be executed.

[0033] If the drone does not reach the designated state, it will continue flying.

[0034] Step S1202: Predict the energy at the parachute opening point.

[0035] The prediction of the parachute opening point energy specifically includes the following sub-steps:

[0036] Step S12021: Obtain the set parachute opening point energy.

[0037] Specifically, the set parachute deployment point energy is pre-calculated. At this point, the drone has not actually reached the deployment point, and the calculated deployment point energy... V0 is the preset parachute deployment speed, and m represents the mass of the drone.

[0038] Step S12022: In response to obtaining the set opening point energy, calculate the current point energy.

[0039] Specifically, the energy of the drone at its current location is calculated, where the energy at the current location is Ev = 0.5mV. 2 V represents the current flight speed, and m represents the mass of the drone.

[0040] Step S12023: In response to calculating the energy at the current point, calculate the height energy difference between the current point and the recovery point.

[0041] Specifically, the calculation involves the height energy difference Eh = mg(h - h0) between the current point of the drone and the recovery point, where h is the current height of the drone, h0 is the set recovery height, g represents the gravitational acceleration, and m represents the mass of the drone.

[0042] Furthermore, the parameter g in the formula Eh = mg(h - h0) can be corrected according to the actual flight altitude range.

[0043] The specific correction method is as follows: if the difference between the current altitude of the drone and the set recovery altitude (h-h0) > 10000m, then g = Kz * g0, where Kz = 0.98 and g0 = 9.8.

[0044] If the difference between the current altitude of the drone and the set recovery altitude (h-h0) is less than 10000m, then g = Kz * g0, Kz = 1, g0 = 9.8.

[0045] Step S12024: In response to calculating the altitude energy difference between the current point and the recovery point, calculate the aerodynamic energy loss difference between the current point and the recovery point of the UAV.

[0046] Among them, the aerodynamic energy loss difference between the current point and the recovery point of the drone Where ΔR is the distance to be flown and k is the correction factor. ρ is the air density, S is the characteristic area, and C is the characteristic area. D Here are the aerodynamic parameters: V is the current flight speed, and V0 is the set parachute deployment point speed.

[0047] Furthermore, based on the actual speed, altitude range, and simulation results, the aerodynamic energy loss difference can be assessed. and In the formula, k and C D Perform parameter correction.

[0048] Where for C D The correction needs to take into account the high angle of attack during the recovery phase, and can generally be adjusted according to the actual situation. D Correct by multiplying by a scaling factor of 1.2;

[0049] The value of k can be corrected based on the simulation results and prediction algorithm. The value of k can be taken in the range of 0.9 to 1.2 so that the aerodynamic energy loss difference Ea between the current point of the UAV and the recovery point is less than 100.

[0050] Step S12025: Predict the parachute opening point energy based on the current point energy, altitude energy difference, and aerodynamic energy loss difference.

[0051] Specifically, the energy at the parachute deployment point, Ey, is calculated as Ev + Eh - Ea, where Ev represents the current energy of the UAV, Eh represents the energy difference in altitude between the current point and the recovery point, and Ea represents the aerodynamic energy loss difference between the current point and the recovery point.

[0052] Step S130: Issue a parachute deployment command based on the predicted deployment point energy, and complete the recovery by reaching the preset position according to the deployment command.

[0053] Step S130 specifically includes the following sub-steps:

[0054] Step S1301: Compare the predicted parachute opening point energy with the set parachute opening point energy.

[0055] Step S1302: Issue a parachute opening command based on the comparison results.

[0056] Specifically, the predicted parachute deployment point energy Ey is compared with the set parachute deployment point energy E0. When the comparison result is Ey≤E0, a parachute deployment command is issued. The parachute deployment command instructs the engine to shut down and the drone to wait for its flight speed to drop to V0 or below before flying to the parachute deployment point to deploy the parachute. This allows the drone to be recovered when it reaches the recovery point.

[0057] The above-described recycling strategy in this embodiment, compared with other recycling strategies in practical applications, has a recycling location control accuracy better than 30m and a recycling speed control accuracy better than 0.1m / s, and its overall effect is far superior to existing recycling strategies.

[0058] Example 2

[0059] like Figure 2 The image shows an embodiment of an energy prediction-based drone recovery system provided in this application. The drone recovery system can communicate remotely with the drone to complete the drone recovery operation, or it can be installed inside the drone so that the drone can perform the recovery operation autonomously.

[0060] The drone recovery system based on energy prediction specifically includes: a setting unit 310, a prediction unit 320, and a command issuing unit 330.

[0061] The setting unit 310 sets the flight path.

[0062] During the route setting process, the route includes multiple waypoints and parachute deployment points.

[0063] The prediction unit 320 is connected to the setting unit 310 and is used to predict the energy of the parachute opening point according to the set flight path.

[0064] During the flight of the UAV according to the set route, the prediction unit 320 predicts the point at which the UAV can deploy its parachute.

[0065] Specifically, the prediction unit 320 includes the following modules: a judgment module and a prediction module.

[0066] The judgment module is used to determine whether the drone has reached a specified state.

[0067] The prediction module is connected to the judgment module and is used to predict the energy of the parachute deployment point if the UAV reaches a specified state.

[0068] If the drone does not reach the designated state, it will continue flying.

[0069] Specifically, the prediction module includes a parachute opening point energy acquisition module, a current energy acquisition module, an altitude energy difference acquisition module, an aerodynamic energy loss difference acquisition module, and a parachute opening point energy prediction module during the parachute opening point energy prediction process.

[0070] The parachute deployment point energy acquisition module is used to acquire the set parachute deployment point energy.

[0071] The current energy acquisition module is connected to the parachute opening point energy acquisition module and is used to calculate the energy at the current point.

[0072] The height energy difference acquisition module is connected to the current energy acquisition module and is used to calculate the height energy difference between the current point and the recovery point.

[0073] The aerodynamic energy loss difference acquisition module is connected to the altitude energy difference acquisition module and is used to calculate the aerodynamic energy loss difference between the current point of the UAV and the recovery point.

[0074] The parachute deployment point energy prediction module is connected to the current energy acquisition module, the altitude energy difference acquisition module, and the aerodynamic energy loss difference acquisition module, respectively, and is used to predict the parachute deployment point energy based on the current point energy, the altitude energy difference, and the aerodynamic energy loss difference.

[0075] The command issuing unit 330 is connected to the prediction unit 320 and is used to issue a parachute opening command based on the predicted parachute opening point energy, and to complete the recovery by reaching a preset position according to the parachute opening command.

[0076] Specifically, the instruction issuing unit 330 includes the following sub-modules: a comparison module and an umbrella opening instruction issuing module.

[0077] The comparison module is used to compare the predicted parachute deployment point energy with the set parachute deployment point energy.

[0078] When the comparison result Ey≤E0, the umbrella opening command issuing module issues an umbrella opening command.

[0079] Example 3

[0080] This embodiment specifically proposes a drone, including the drone body and a drone recovery system based on energy prediction. Specifically, it describes the scheme in Embodiment 2 where the drone recovery system based on energy prediction is installed inside the drone.

[0081] The drone itself can be any type of drone currently in use.

[0082] The energy prediction-based drone recovery system is installed in the drone and includes a setting unit, a prediction unit, and a command issuing unit. The setting unit pre-sets the flight path, and the prediction unit and command issuing unit control the drone to perform the following steps based on the pre-set flight path:

[0083] The prediction unit predicts the energy at the parachute deployment point based on the set flight path.

[0084] The command issuing unit issues a parachute deployment command based on the predicted deployment point energy, and completes the recovery by reaching the preset position according to the deployment command.

[0085] The prediction of the parachute deployment point energy based on the set flight path specifically includes:

[0086] Determine whether the specified state has been reached.

[0087] The determination of whether a specified state has been reached includes determining whether the drone has reached a specified location or a specified speed. If the drone has reached a specified location or a specified speed, then step S12 is executed.

[0088] If the drone does not reach the designated state, it will continue flying.

[0089] For example, if the drone arrives at waypoint Xn-1, the energy at the parachute deployment point is predicted.

[0090] The prediction of the parachute opening point energy specifically includes the following sub-steps:

[0091] Step S1: Obtain the set parachute opening point energy.

[0092] Specifically, the set parachute deployment point energy is pre-calculated. At this point, the drone has not actually reached the deployment point, and the calculated deployment point energy... V0 is the set parachute deployment speed.

[0093] Step S2: In response to obtaining the set opening point energy, calculate the current point energy.

[0094] Specifically, the energy of the drone at its current location is calculated, where the energy at the current location is Ev = 0.5mV. 2 V represents the current flight speed, and m represents the mass of the drone.

[0095] Step S3: In response to calculating the energy at the current point, calculate the height energy difference between the current point and the recovery point.

[0096] Specifically, the calculation involves the height energy difference Eh = mg(h - h0) between the current point of the drone and the recovery point, where h is the current height of the drone, h0 is the set recovery height, g represents the gravitational acceleration, and m represents the mass of the drone.

[0097] Furthermore, the parameter g in the formula Eh = mg(h - h0) can be corrected according to the actual flight altitude range.

[0098] The specific correction method is as follows: if the difference between the current altitude of the drone and the set recovery altitude (h-h0) > 10000m, then g = Kz * g0, where Kz = 0.98 and g0 = 9.8.

[0099] If the difference between the current altitude of the drone and the set recovery altitude (h-h0) is less than 10000m, then g = Kz * g0, Kz = 1, g0 = 9.8.

[0100] Step S4: In response to the calculation of the altitude energy difference between the current point and the recovery point, calculate the aerodynamic energy loss difference between the current point and the recovery point of the UAV.

[0101] Among them, the aerodynamic energy loss difference between the current point and the recovery point of the drone Where ΔR is the distance to be flown and k is the correction factor. ρ is the air density, S is the characteristic area, and C is the characteristic area. D Here are the aerodynamic parameters: V is the current flight speed, and V0 is the set parachute deployment point speed.

[0102] Furthermore, based on the actual speed, altitude range, and simulation results, the aerodynamic energy loss difference can be assessed. and In the formula, k and C D Perform parameter correction.

[0103] Where for C D The correction needs to take into account the high angle of attack during the recovery phase, and can generally be adjusted according to the actual situation. D Correct by multiplying by a scaling factor of 1.2;

[0104] The value of k can be corrected based on the simulation results and prediction algorithm. The value of k can be taken in the range of 0.9 to 1.2 so that the aerodynamic energy loss difference Ea between the current point of the UAV and the recovery point is less than 100.

[0105] Step S5: Predict the parachute opening point energy based on the current point energy, altitude energy difference, and aerodynamic energy loss difference.

[0106] Specifically, the energy at the parachute deployment point, Ey, is calculated as Ev + Eh - Ea, where Ev represents the current energy of the UAV, Eh represents the energy difference in altitude between the current point and the recovery point, and Ea represents the aerodynamic energy loss difference between the current point and the recovery point.

[0107] Furthermore, the predicted parachute deployment point energy Ey is compared with the set parachute deployment point energy E0. When the comparison result is Ey≤E0, a parachute deployment command is issued. The parachute deployment command instructs the engine to shut down and the UAV to wait for its flight speed to drop to V0 or below before flying to the parachute deployment point to deploy the parachute, thereby completing the recovery of the UAV in the recovery range area.

[0108] By deploying an energy-prediction-based drone recovery system in drones, the drones can calculate various energy parameters and, combined with the control of engine switching under specified conditions, autonomously complete the recovery process, thus improving the accuracy of drone recovery.

[0109] This application has the following beneficial effects:

[0110] The UAV recovery method proposed in this application can effectively improve the accuracy of the UAV's parachute deployment point position and speed. Compared with other recovery strategies in the prior art, the UAV recovery method proposed in this application achieves a recovery position control accuracy better than 30m and a recovery speed control accuracy better than 0.1m / s, with significant improvements in both control accuracy and recovery speed. The overall effect is far superior to existing recovery strategies.

[0111] Although the examples referenced in this application are described for illustrative purposes only and not for limiting the scope of this application, changes, additions and / or deletions to the implementation may be made without departing from the scope of this application.

[0112] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for recovering unmanned aerial vehicles (UAVs) based on energy prediction, characterized in that, Specifically, the following steps are included: Set the flight route; Based on the set flight path, predict the energy at the parachute deployment point; Based on the predicted energy at the parachute deployment point, a parachute deployment command is issued, and the parachute is retrieved at the preset location according to the deployment command. Predicting the energy at the parachute deployment point based on the set flight path includes the following sub-steps: Determine whether a specified state has been reached; If the drone reaches the designated state, the energy at the parachute deployment point will be predicted. If the drone does not reach the designated state, it will continue flying; The prediction of the parachute opening point energy includes the following sub-steps: Obtain the energy at the set parachute deployment point; In response to obtaining the set parachute deployment point energy, calculate the energy at the current point; In response to calculating the energy at the current point, calculate the height energy difference between the current point and the recovery point; In response to the calculation of the altitude energy difference between the current point and the recovery point, the aerodynamic energy loss difference between the current point and the recovery point of the UAV is calculated. Aerodynamic energy loss difference between the current point and the recovery point of the drone Where ΔR is the distance to be flown and k is the correction factor. ρ is the air density, S is the characteristic area, and C is the characteristic area. D Here are the aerodynamic parameters: V is the current flight speed, and V0 is the set parachute deployment point speed. Based on the actual speed, altitude range, and simulation results, the aerodynamic energy loss difference is... and In the formula, k and C D Perform parameter corrections to ensure that the aerodynamic energy loss difference Ea between the current point and the recovery point of the UAV is less than 100. The parachute deployment point energy is predicted based on the current point energy, the energy difference at altitude, and the aerodynamic energy loss difference.

2. The UAV recovery method based on energy prediction as described in claim 1, characterized in that, The route includes multiple waypoints, parachute deployment points, and recovery points.

3. The UAV recovery method based on energy prediction as described in claim 1, characterized in that, Determining whether a specified state has been reached includes determining whether the drone has reached a specified location or whether the drone's flight speed has reached a specified flight speed.

4. The UAV recovery method based on energy prediction as described in claim 3, characterized in that, Predicting the deployment point energy based on the set flight path includes obtaining the set deployment point energy.

5. The UAV recovery method based on energy prediction as described in claim 4, characterized in that, The set parachute deployment point energy is specifically expressed as follows: Where V0 is the set parachute deployment speed, and m represents the mass of the drone.

6. A drone recovery system based on energy prediction, characterized in that, Specifically, it includes: Setting unit and instruction issuing unit; The setting unit is used to set the flight path; The prediction unit is used to predict the energy at the parachute deployment point based on the set flight path. The command issuing unit is used to issue a parachute opening command based on the predicted parachute opening point energy, and to complete the recovery by reaching the preset position according to the parachute opening command. The prediction unit predicts the energy at the parachute deployment point based on the set flight path, including the following sub-steps: Determine whether a specified state has been reached; If the drone reaches the designated state, the energy at the parachute deployment point will be predicted. If the drone does not reach the designated state, it will continue flying; The prediction of the parachute opening point energy includes the following sub-steps: Obtain the energy at the set parachute deployment point; In response to obtaining the set parachute deployment point energy, calculate the energy at the current point; In response to calculating the energy at the current point, calculate the height energy difference between the current point and the recovery point; In response to the calculation of the altitude energy difference between the current point and the recovery point, the aerodynamic energy loss difference between the current point and the recovery point of the UAV is calculated. Aerodynamic energy loss difference between the current point and the recovery point of the drone Where ΔR is the distance to be flown and k is the correction factor. ρ is the air density, S is the characteristic area, and C is the characteristic area. D Here are the aerodynamic parameters: V is the current flight speed, and V0 is the set parachute deployment point speed. Based on the actual speed, altitude range, and simulation results, the aerodynamic energy loss difference is... and In the formula, k and C D Perform parameter corrections to ensure that the aerodynamic energy loss difference Ea between the current point and the recovery point of the UAV is less than 100. The parachute deployment point energy is predicted based on the current point energy, the energy difference at altitude, and the aerodynamic energy loss difference.

7. The UAV recovery system based on energy prediction as described in claim 6, characterized in that, The prediction unit specifically includes a judgment module and a prediction module; The judgment module is used to determine whether the drone has reached a specified state; If the drone reaches the designated state, the prediction module will predict the energy at the parachute deployment point. If the drone does not reach the designated state, it will continue flying.

8. The UAV recovery system based on energy prediction as described in claim 7, characterized in that, The prediction module includes a parachute deployment point energy acquisition module, which acquires the set parachute deployment point energy.

9. A drone, characterized in that, Specifically, it includes the drone itself, and the drone recovery system based on energy prediction as described in any one of claims 6-8.

Images