JP2012168037A - Velocity prediction device and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To enable a velocity pattern to be accurately predicted even for a road on which a user does not travel before.SOLUTION: A road gradient acquisition section 37 acquires a road gradient of each link along a search route. An operation state prediction section 38 predicts time-series data on an acceleration operation state at each link based on a state transition probability of the acceleration operation state learnt for each state of road gradient as well as the acquired road gradient. A velocity pattern generation section 40 predicts time-series data on velocity at each link based on a relation between the road gradient learnt for each acceleration operation state and the acceleration, the acquired road gradient and the predicted time-series data on the acceleration operation state.

Description

  The present invention relates to a speed predicting apparatus and a program, and more particularly, to a speed predicting apparatus and a program for predicting time-series data of the speed of a vehicle on a traveling road.

  Conventionally, travel data and travel environment data are stored in association with each other, travel speed pattern candidates are generated based on the travel data, travel speed pattern candidates that match the current travel environment data are extracted, and travel is performed from now on. A traveling speed pattern estimation device that outputs an estimated traveling speed pattern of a route to travel is known (Patent Document 1).

JP 2004-282848 A

  However, with the technique described in Patent Document 1, it is necessary to accumulate past travel data as a database, and it is impossible to accurately predict a speed pattern for a road that has not traveled in the past. There is.

  The present invention has been made to solve the above problems, and provides a speed prediction device and program capable of accurately predicting a speed pattern even on a road that has never traveled in the past. With the goal.

  In order to achieve the above object, a speed prediction apparatus according to the present invention includes a road state acquisition means for acquiring a road state at each point of a prediction target travel path on which a vehicle travels, and a plurality of types of roads for the driver of the vehicle. Based on the state transition probabilities of a plurality of types of operation states when the driver operates the vehicle and the road state of each point acquired by the road state acquisition unit, which is obtained in advance for each state, Operation state prediction means for predicting time series data of the operation state, a relationship between the road state and the acceleration of the vehicle obtained in advance for each of the plurality of types of operation states, and each point acquired by the road state acquisition means Time series data of acceleration of the vehicle on the travel road based on the road state of the vehicle and the time series data of the operation state predicted by the operation state prediction means Predicted, on the basis of the time series data of acceleration of the vehicle in the traveling path, is configured to include a, a speed estimation means for estimating the time-series data of the velocity of the vehicle in the traveling path.

  The program according to the present invention is such that the computer is preliminarily determined for each of a plurality of types of road conditions for road condition acquisition means for acquiring road conditions at each point of a prediction target travel path on which the vehicle travels, and for the driver of the vehicle. Based on the state transition probabilities of the plurality of types of operation states when the driver operates the vehicle and the road state of each point acquired by the road state acquisition means, the time series data of the operation states on the travel road The operation state prediction means for predicting, the relationship between the road state obtained in advance for each of the plurality of types of operation states and the acceleration of the vehicle, the road state of each point acquired by the road state acquisition means, and the operation state Based on the time series data of the operation state predicted by the prediction means, the time series data of the acceleration of the vehicle on the travel path is predicted, Based on the time series data of acceleration of the vehicle in the traveling road is a program for functioning as a speed prediction means for predicting time series data of the velocity of the vehicle in the traveling path.

  According to the present invention, the road condition acquisition means acquires the road condition of each point on the predicted travel path on which the vehicle travels. By the operation state prediction means, the state transition probabilities of the plurality of types of operation states obtained by the driver operating the vehicle in advance for each of the plurality of types of road conditions for the driver of the vehicle, and each point acquired by the road state acquisition means The time series data of the operation state on the traveling road is predicted based on the road state.

  Then, the speed prediction means predicts the relationship between the road state and the vehicle acceleration obtained in advance for each of a plurality of types of operation states, the road state of each point acquired by the road state acquisition means, and the operation state prediction means. The time series data of the acceleration of the vehicle on the road is predicted based on the time series data of the operation state, and the time series of the speed of the vehicle on the road based on the time series data of the acceleration of the vehicle on the road Predict data.

  As described above, the time series data of the operation state is predicted based on the state transition probability of the operation state obtained in advance for each road state for the driver of the vehicle, and the road state and the acceleration of the vehicle obtained in advance for each operation state are predicted. By predicting the time series data of the acceleration of the vehicle using the relationship, the speed pattern can be accurately predicted even on a road that has not traveled in the past.

  According to the present invention, the relationship between the road state for each of the plurality of types of operation states and the acceleration of the vehicle can be obtained in advance for the driver to be predicted. As a result, the speed pattern can be predicted with higher accuracy.

  The speed prediction apparatus according to the present invention further includes energy amount calculation means for calculating the amount of energy consumed when the vehicle travels on the travel path based on the time-series data of the speed predicted by the speed prediction means. Can do. As a result, the amount of energy in the travel path can be accurately predicted.

  In addition, the energy amount calculation means described above includes the time series data of the speed predicted when the operation state prediction means repeatedly predicts the operation state time series data and the speed prediction means predicts the speed time series data. For each, the amount of energy can be calculated, and the range of the amount of energy consumed when traveling on the road can be calculated. As a result, the amount of energy on the travel path can be accurately predicted in consideration of the fluctuation range.

  The program of the present invention can also be provided by being stored in a storage medium.

  As described above, according to the speed prediction apparatus and program of the present invention, the time series data of the operation state is predicted based on the state transition probability of the operation state obtained in advance for each road state for the driver of the vehicle, and the operation By predicting the time series data of the vehicle acceleration using the relationship between the road condition and the vehicle acceleration obtained in advance for each state, the speed pattern can be accurately obtained even on a road that has never traveled in the past. The effect that it can predict is acquired.

It is a block diagram which shows the route search apparatus which concerns on the 1st Embodiment of this invention. (A) The figure which shows the state transition probability of the accelerator operation state with respect to the down road gradient state, (B) The figure which shows the state transition probability of the accelerator operation state with respect to the flat road slope state, and (C) The up road gradient It is a figure which shows the state transition probability of the accelerator operation state with respect to this state. It is a graph showing the relationship between the road gradient and acceleration for each of the three accelerator operating states. It is a flowchart which shows the content of the driving characteristic learning process routine in the route search apparatus which concerns on the 1st Embodiment of this invention. It is a flowchart which shows the content of the route search process routine in the route search apparatus which concerns on the 1st Embodiment of this invention. It is a flowchart which shows the content of the actual travel energy calculation process routine in the route search apparatus which concerns on the 1st Embodiment of this invention. (A) The graph which shows the time series data of the produced | generated speed pattern and the actual value of speed, (B) The graph which shows the detail of the time series data of the produced | generated speed pattern and the actual value of speed. It is a graph which shows the calculation result of traveling energy amount. It is a block diagram which shows the route search apparatus which concerns on the 1st Embodiment of this invention.

  Hereinafter, preferred embodiments of the present invention will be described with reference to the drawings. An example will be described in which the present invention is applied to a route search device that is mounted on a vehicle and guides a route to a destination for a driver of the host vehicle.

  As shown in FIG. 1, the route search apparatus 10 according to the first embodiment includes an operation amount sensor 12 that detects an operation amount of an accelerator pedal, a vehicle speed sensor 13 that detects the vehicle speed of the host vehicle, An acceleration sensor 14 that detects acceleration, a position measurement unit 16 that measures the position of the host vehicle, an input operation unit 18 for a driver to input a destination, and a route to a destination that optimizes the amount of travel energy A computer 20 for searching, a display device 22 for displaying the searched route to the driver, and an information presentation unit 24 for presenting to the driver that the traveling was less than the predicted traveling energy amount. I have.

  The operation amount sensor 12 detects the accelerator opening as the operation amount of the accelerator pedal.

  The position measurement unit 16 is configured using, for example, a GPS sensor, and measures the current position of the host vehicle.

  The information presentation unit 24 is configured using, for example, an indicator, and presents to the driver that the travel is less than the predicted travel energy amount by lighting the indicator.

  The computer 20 includes a CPU, a RAM, and a ROM that stores programs for executing a driving characteristic learning process routine, a route search process routine, and an actual travel energy calculation process routine, which will be described later. It is configured as shown. The computer 20 includes a road network database 26 storing road network data and road information indicating road gradients and speed limits of each link on the road network, an accelerator pedal operation amount detected by the operation amount sensor 12, and a vehicle speed sensor. The own vehicle that acquires the vehicle speed detected by the vehicle 13, the acceleration of the vehicle detected by the acceleration sensor 14, the position of the vehicle measured by the position measurement unit 16, and the road gradient at the position as vehicle information. Based on the information acquisition unit 28, the acquired own vehicle information, the driving characteristic learning unit 30 that learns the state transition probability of the accelerator operation state as the operation characteristic, and the relationship between the road gradient and the acceleration as the acceleration characteristic, and learning The operation characteristic storage unit 32 for storing the state transition probability of the accelerator operation state that has been And a acceleration characteristic storage unit 34 that stores the relationship between the road gradient and the acceleration.

  The host vehicle information acquisition unit 28 learns the operation characteristics and acceleration characteristics of the driver for the driver of the host vehicle, the accelerator pedal operation amount detected by the operation amount sensor 12, and the vehicle speed of the host vehicle detected by the vehicle speed sensor 13. Then, the acceleration of the host vehicle detected by the acceleration sensor 14, the position of the host vehicle measured by the position measuring unit 16, and the road gradient at the position obtained from the road network database 26 are sequentially acquired and stored as host vehicle information. .

  The driving characteristic learning unit 30 classifies the accumulated vehicle information, for example, for each three-state road gradient, and for each of the three-state road gradients, as shown in FIGS. The state transition probability of the three-state accelerator operation state is learned. The three-state road gradient is, for example, downward (a road gradient of less than -1%), flat (a road gradient of -1% or more and 1% or less), and upward (a road gradient of more than 1%). The three accelerator operation states are, for example, no accelerator operation (accelerator opening = 0%), small accelerator operation (0% <accelerator opening <20%), and large accelerator operation (20% ≦ accelerator opening). ).

  The driving characteristic learning unit 30 obtains the state transition probability Pr represented by the following expressions (1) to (3) for each of the three-state road gradients as the learning result of the operation characteristic for the driver of the own vehicle. Get.

Here, acc _n represents an access operation state. grad _n represents the state of the road gradient, up represents up, level represents flat, and down represents down.

The driving characteristic learning unit 30 classifies the accumulated own vehicle information, for example, for each of the three accelerator operation states, and as shown in FIG. a road gradient road _n [%] and finding a function that represents the relationship between the acceleration _{a n [km / h / sec} ].

Here, large represents an access operation large, small represents a small accelerator operation, and nop represents no accelerator operation. Further, c _large , c _small , c _nop , b _large , b _small , and b _nop are parameters that determine the above functions. b _large , b _small , and b _nop are not necessarily constants. In order to express the variation in acceleration in FIG. 3, the parameters may be obtained by maximum likelihood estimation or the like using b _large , b _small , and b _nop as random variables. Specifically, σ may be estimated from the stored vehicle information by using b as a random variable according to a normal distribution having an average of 0 and a variance σ ² .

  The function representing the relationship between the road gradient and the acceleration is not limited to the linear expression as shown in the above expression (4), but is expressed as a secondary expression as shown in the following expression (5). It may be.

However, d _large , d _small , and d _nop are parameters that define the above function.

  The operation characteristic storage unit 32 stores the state transition probability of the accelerator operation state represented by the above equations (1) to (3) learned for each road gradient state. Further, the acceleration characteristic storage unit 34 stores the parameters of the function represented by the above equation (4) learned for each accelerator operation state.

  Further, the computer 20 determines the destination from the departure point based on the destination input by the operation of the input operation unit 18 by the driver, the position of the host vehicle as the departure point measured by the position measurement unit 16, and the road network data. A route candidate search unit 36 that searches a plurality of route candidates to the ground, a road gradient acquisition unit 37 that acquires a road gradient from the road network database 26 for each link included in the searched plurality of route candidates, and a search For each link included in a plurality of route candidates, an operation state prediction unit 38 that predicts time series data of the accelerator operation state based on the road gradient of the link and the learned state transition probability of the accelerator operation state, and a search Time-series data of predicted accelerator operation status, learning A speed pattern generation unit 40 that predicts acceleration time-series data and generates a speed pattern based on a function representing the relationship between the road gradient and acceleration and the road gradient of the link, and a plurality of searched route candidates A travel energy calculation unit 42 that calculates a travel energy amount consumed when traveling based on the generated speed pattern, time series data of predicted acceleration, and road gradient of the link for each link included in the link. And a minimum energy route selection unit 44 that selects a route that minimizes the total amount of travel energy from the searched plurality of route candidates and outputs the selected route to the display device 22. The speed pattern generation unit 40 is an example of a speed prediction unit.

  The route candidate search unit 36 uses an algorithm for searching for the shortest route based on the destination input by the operation of the input operation unit 18, the departure place measured by the position measurement unit 16, and road network data, Using a route search algorithm or the like, a plurality of route candidates from the departure point to the destination are searched. As the route search algorithm, a conventionally known algorithm may be used, and detailed description thereof is omitted.

  For each link included in the plurality of searched route candidates, the operation state prediction unit 38 acquires the road gradient of the link acquired by the road gradient acquisition unit 37 and the accelerator operation state learned for each road gradient state. Based on the state transition probability, time-series data of the accelerator operation state of the link is predicted as described below. Each link included in a plurality of route candidates indicates each of all links included in at least one of the plurality of route candidates.

First, as an initial state of the target link, the initial value of the state of the initial value S ₀ and the road gradient of the accelerator operation state road ₀ (state of the road gradient based on the road gradient of the obtained the link). Then, a random number is generated, and the accelerator at the next time is stochastically using the generated random number in accordance with the state transition probability Pr (S ₁ | S ₀ , load ₀ ) of the accelerator operation state with respect to the current road gradient state load ₀ . to select the operation state S _1. Similarly, the stochastic selection of the accelerator operation state S _{n + 1 at} the next time using the state transition probability Pr (S _{n + 1} | S _n , load _n ) of the accelerator operation state with respect to the current road gradient state load _n Is repeated until the end of the link is reached, and time series data _Sn (n = 0,..., N) in the accelerator operating state is predicted.

For each link included in the plurality of searched route candidates, the speed pattern generation unit 40 loads the road gradient state load _n (n = 0,...) Based on the road gradient of the link acquired by the road gradient acquisition unit 37. , N) and time series data acc _n (n = 0,..., N) of the accelerator steering state predicted for the link, It represented, in accordance with a function with respect to the accelerator operation state acc _n, the time series data _{a n (n = 0, ...} , n) of the acceleration to predict.

The speed pattern generating section 40, for each link included in the searched plurality of path candidates, the time series data a _n predicted acceleration for the link (n = 0, ..., N ) based on according to the following equation (6), as the speed pattern, the time-series data _{v n (n = 0, ...} , n) of the velocity is calculated.

  The initial speed v0 may be the final speed calculated for the previous link or the speed limit of the link obtained from the road network database 26.

For each link included in the searched plurality of route candidates, the travel energy calculation unit 42 calculates time-series data v _n (n = 0,..., N) of the speed calculated for the link and the link. time series data a _n predicted acceleration (n = 0, ..., n) based on the, according to the following equation (7), calculate the traveling energy amount J consumed in traveling the link To do.

Where μ is the rolling friction coefficient, M is the vehicle mass [kg], and g is the gravitational acceleration. к is a resistance coefficient [kg / m], θ is a road gradient (with positive ascending), and m is a partial equivalent mass [kg] during partial acceleration. α is the vehicle acceleration [m / s ^2], obtained from the time series data a _n predicted acceleration for the link. Further, the integration range in the above equation (7) is a region where the driving force F is greater than zero.

  A series of processes including prediction of time series data of an accelerator operation state by the operation state prediction unit 38, generation of a speed pattern by the speed pattern generation unit 40, and calculation of a travel energy amount by the travel energy calculation unit 42 are repeatedly performed, and search is performed. A plurality of travel energy amounts are calculated for each link included in the plurality of route candidates.

  The travel energy calculation unit 42 obtains a range of travel energy amounts obtained from a plurality of travel energy amounts calculated based on a plurality of random number sequences for each link included in the searched plurality of route candidates. .

  The minimum energy route selection unit 44 calculates, for each of the plurality of searched route candidates, the sum of the range of travel energy amounts calculated for each link included in the route candidate, and calculates the total travel energy in the route candidate. Calculate the range of quantities. Further, the minimum energy route selection unit 44 selects a route candidate having the minimum median value of the total travel energy amount range, and displays the route candidate together with the range of the total travel energy amount of the selected route candidate. 22 to output.

  In addition, the computer 20 accumulates the own vehicle information acquired until it arrives at the destination, and is selected by the actual travel energy calculation unit 46 that calculates the total travel energy amount in actual travel and the minimum energy route selection unit 44. The comparison unit 48 further compares the lower limit of the range of the total travel energy amount of the route candidates and outputs the comparison result by the information presentation unit 24.

  The actual running energy calculation unit 46 accumulates the vehicle information acquired by the vehicle information acquisition unit 28 from the departure point to the destination on the searched route, and the vehicle speed, acceleration, Based on the road gradient, the total travel energy amount in actual travel is calculated according to the above equation (7).

  The comparison unit 48 compares the lower limit of the range of the total travel energy amount of the route candidate selected by the minimum energy route selection unit 44 with the total travel energy amount calculated by the actual travel energy calculation unit 46, and If the actual total travel energy amount is smaller than the lower limit of the total travel energy amount range, the indicator of the information presentation unit 24 is turned on to indicate that the travel is less than the predicted travel energy amount. To inform.

  Next, the operation of the route search apparatus 10 according to the first embodiment will be described. First, in the route search device 10, when the host vehicle is traveling by the driving operation of the host vehicle driver, a driving characteristic learning process routine shown in FIG. 4 is executed in the computer 20.

  In step 100, the detected values of the operation amount sensor 12, the vehicle speed sensor 13, and the acceleration sensor 14 and the current position of the host vehicle measured by the position measuring unit 16 are acquired as host vehicle information. In step 102, it is determined whether or not a predetermined amount of the vehicle information acquired in step 100 is accumulated. If the vehicle information is not accumulated in a predetermined amount, the process returns to step 100. On the other hand, if the vehicle information has accumulated a predetermined amount, the process proceeds to step 104.

  In step 104, the accumulated vehicle information is classified for each of the three road gradients, and the state of the accelerator operation state is determined based on the operation amount of the accelerator pedal of the classified vehicle information for each road gradient state. The transition probability is learned, and the state transition probability of the accelerator operation state for each of the three states of the road gradient is stored in the operation characteristic storage unit 32.

  In step 106, the accumulated vehicle information is classified for each of the three accelerator operation states, and for each accelerator operation state, based on the road gradient and acceleration of the vehicle information classified into the accelerator operation state. Then, a function representing the relationship between the road gradient and the acceleration is learned, the parameter of the function learned for each accelerator operation state is stored in the acceleration characteristic storage unit 34, and the driving characteristic learning processing routine is finished. .

  In addition, the computer 20 executes the above-described driving characteristic learning processing routine every predetermined period. Each time it is executed, the computer 20 adds the own vehicle information, updates the accumulated own vehicle information, and updates the updated own vehicle information. Based on the vehicle information, the state transition probability of the accelerator operation state with respect to each of the three states of the road gradient and the function representing the relationship between the road gradient and the acceleration with respect to each of the accelerator operation states are learned, and the learning result is updated.

  When the driver operates the input operation unit 18 to input a destination, the route search processing routine shown in FIG.

  First, in step 110, the current position of the host vehicle measured by the position measuring unit 16 is acquired, the acquired current position of the host vehicle is set as a departure point, and the destination input by the driver is indicated. Get information and set destination.

  In step 112, a plurality of route candidates from the departure point to the destination are searched by the route search algorithm. In the next step 114, one link to be predicted is selected from the links included in the plurality of route candidates searched in step 112. In step 115, information indicating the road gradient of the link to be predicted is acquired from the road network database 26. In step 116, a variable n indicating the number of repetitions is set to an initial value of 1. In step 118, time-series data of the driver's accelerator operation state is predicted for the prediction target link.

  In step 120, acceleration time-series data is predicted for the prediction target link using the accelerator operation state time-series data predicted in step 114, and a speed pattern is generated.

  In the next step 122, based on the speed pattern generated in step 120, the time series data of acceleration, and the road gradient of the prediction target link obtained from the road network database 26, the travel energy in the prediction target link is calculated. Calculate the amount.

  In step 124, it is determined whether or not the variable n has reached the maximum value N of the number of repetitions. If the variable n has not reached N, in step 126, the variable n is incremented and the process returns to step 118. If the variable n reaches N, the process proceeds to step 128. At this time, N travel energy amounts are obtained for the prediction target link.

  In step 128, it is determined whether or not the processing in steps 116 to 126 has been performed for all links included in the plurality of route candidates searched in step 112, and the processing in steps 116 to 126 is performed. When there is no link, the process returns to step 114, and the link is set as a prediction target link.

  On the other hand, if the processing in steps 116 to 126 has been performed for all the links included in the plurality of route candidates searched in step 112, the process proceeds to step 130, and the plurality of routes searched in step 112. For each candidate, the range of total travel energy is calculated.

  In step 132, a route candidate that minimizes the total travel energy amount calculated in step 130 is selected. In step 134, the selected route candidate and the total travel energy amount calculated for the route candidate are selected. The range is output to the display device 22, and the route search processing routine is terminated.

  When the route search processing routine is executed, for example, in a vehicle-mounted navigation system, processing for guiding the driver to the searched route to the destination and the range of the total travel energy amount is performed.

  In addition, after the route search processing routine is executed, the actual running energy calculation processing routine shown in FIG.

  First, in step 140, the detected values of the operation amount sensor 12, the vehicle speed sensor 13, and the acceleration sensor 14 and the current position of the host vehicle measured by the position measuring unit 16 are acquired as host vehicle information. In step 142, it is determined whether or not the current position of the host vehicle in the host vehicle information acquired in step 100 has reached the destination set in the route search processing routine. If not, the process returns to step 140. On the other hand, if the current position of the host vehicle has reached the set destination, the process proceeds to step 144.

  In step 144, the total travel energy amount in actual travel to the destination is calculated based on the own vehicle information accumulated until the destination is reached. In step 146, whether or not the actual total travel energy amount calculated in step 144 is smaller than the lower limit of the total travel energy amount range calculated for the route candidate selected in the route search processing routine. judge. When the calculated actual total travel energy amount is equal to or greater than the lower limit of the calculated total travel energy amount range, the actual travel energy calculation processing routine is terminated. On the other hand, when the calculated actual total travel energy amount is smaller than the lower limit of the calculated total travel energy amount range, in step 148, the indicator of the information presentation unit 24 is turned on, and the actual travel energy calculation process is performed. End the routine.

  Next, experimental results for determining the speed pattern and the travel energy amount by the speed pattern generation method and the travel energy amount calculation method described in the above embodiment will be described.

  When the speed pattern generated by the speed pattern generation method described in the above embodiment is compared with the speed pattern in actual traveling, a result as shown in FIG. 7A is obtained. As shown in FIG. 7B, when the detailed comparison result is seen, the change in the accelerator operation state does not match and the speed pattern is different, but the statistical property (probability of appearance of the accelerator operation state) is It can be seen that it is maintained.

  In addition, the result of calculating the travel energy amount based on the speed pattern in which the speed does not change in the link, the travel energy amount calculated by the travel energy amount calculation method described in the above embodiment, and the actual travel Comparing with the travel energy amount, as shown in FIG. 8, it was found that the travel energy amount can be accurately obtained by the travel energy amount calculation method described in the above embodiment. In addition, it was found that the range of the travel energy amount in consideration of the fluctuation range can be accurately performed.

  As described above, according to the route search device according to the first embodiment, the accelerator for each link based on the state transition probability of the accelerator operation state learned for each road gradient state for the driver of the host vehicle. Predict time series data of operation states and predict time series data of vehicle acceleration for each link using a function representing the relationship between road gradient and vehicle acceleration learned for each accelerator operation state for the driver. Thus, even on a road that has never traveled in the past, the speed pattern of each link can be accurately predicted in consideration of individual driving characteristics. Further, based on the speed pattern of each link, the total travel energy amount on the searched route can be predicted with high accuracy, and the total travel energy amount including the fluctuation range can be predicted. In addition, the shortest route with the total travel energy amount as a cost can be searched with high accuracy.

  Further, since the accelerator operation state is predicted and the speed pattern is generated based on the state probability transition of the accelerator operation state, the appearance frequency (time rate) of each accelerator operation state can be matched to the individual driving characteristics. In addition, since the road gradient is taken into consideration, the error between the travel energy amount calculated based on the predicted speed pattern and the actual travel energy amount can be reduced.

  In addition, instead of generating a speed pattern using past driving data, personal characteristics of accelerator operation (state probability transition of accelerator operation state) and personal acceleration characteristics (road gradient and vehicle acceleration Therefore, the speed pattern can be generated and the travel energy amount can be estimated without the travel data on the road for generating the speed pattern.

  Further, the parameters to be stored are the parameters of the function representing the state transition probability and the acceleration characteristic, and since the number of parameters is small, the mounting is easy and a large amount of data accumulation is unnecessary.

  Next, a second embodiment will be described. In addition, about the part which becomes the structure similar to 1st Embodiment, the same code | symbol is attached | subjected and description is abbreviate | omitted.

  In the second embodiment, in this embodiment, the route search apparatus is mounted on a hybrid vehicle including an internal combustion engine and a motor generator that performs a charging operation and a torque assist operation provided on an output shaft of the internal combustion engine. This is different from the first embodiment in that the charging / discharging schedule is determined based on the amount of travel energy predicted for each link in the route to the destination.

  As shown in FIG. 9, the route search device 210 according to the second embodiment includes an operation amount sensor 12, an acceleration sensor 14, a position measurement unit 16, an input operation unit 18, and a shortest route to a destination. And a computer 220 that determines a charging / discharging operation schedule in the shortest path, a display device 22, and a motor control unit 222 that controls the charging / discharging operation by the motor generator according to the determined charging / discharging operation schedule. I have.

  The motor control unit 222 controls the charging operation and the torque assist operation (discharge operation) of the motor generator provided in the host vehicle that is a hybrid vehicle according to the schedule of the charge / discharge operation determined by the computer 220.

  The computer 220 is input by the operation of the road network database 26, the vehicle information acquisition unit 28, the driving characteristic learning unit 30, the operation characteristic storage unit 32, the acceleration characteristic storage unit 34, and the input operation unit 18 by a driver. A route search unit 236 for searching for the shortest route from the departure point to the destination based on the destination, the position of the host vehicle as the departure point measured by the position measurement unit 16, and the road network data; Based on the travel energy amount of each link in the shortest route searched, the charge / discharge operation schedule in the shortest route is determined based on the unit 37, the operation state prediction unit 38, the speed pattern generation unit 40, the travel energy calculation unit 42. The charging / discharging schedule determination part 244 to determine is provided.

  The route search unit 236 includes an algorithm for searching for the shortest route based on the destination input by the operation of the input operation unit 18, the departure place measured by the position measurement unit 16, and the road network data. The shortest route from the starting point to the destination is searched using an algorithm for searching for and output to the display device 22.

  The operation state prediction unit 38 predicts time series data of the accelerator operation state for each link in the searched shortest path. The speed pattern generation unit 40 predicts acceleration time-series data for each link in the searched shortest path, and generates a speed pattern.

  The traveling energy calculation unit 42 calculates the amount of traveling energy consumed when traveling for each link in the searched shortest route. In the above formula (7), the case where the region where the driving force F is larger than 0 has been described as an example, but in this embodiment, the portion where F <0 is the thermal energy generated during deceleration ( Loss) and energy that can be recovered by power generation (recovered amount). Therefore, considering the regenerative efficiency, the range of F <0 is also integrated.

  Based on the travel energy amount calculated for each link in the searched shortest route, the charging / discharging schedule determination unit 244 determines an optimal operation schedule (charging operation) for each link of the motor generator in the shortest route from the starting point to the destination. , Torque assist operation, no operation) is determined and output to the motor control unit 222. In addition, about the determination method of the schedule of charging / discharging operation | movement, a conventionally well-known method should just be used and detailed description is abbreviate | omitted.

  Note that other configurations and operations of the route search apparatus according to the second embodiment are the same as those of the first embodiment, and thus the description thereof is omitted.

  As described above, according to the route search device according to the second embodiment, the travel energy amount of each link on the searched shortest route can be accurately calculated, and the optimum schedule of the charge / discharge operation can be determined. Can be determined.

  In the first embodiment and the second embodiment described above, the case where the route search device is mounted on a vehicle has been described as an example. However, the present invention is not limited to this, and a device on the traffic information center side. In addition, the present invention may be applied. In this case, the vehicle information collected from the vehicle to be searched for the route and the data indicating the starting point and the destination are transmitted to the device on the traffic information center side. The route search is performed and the route search result (travel route and total travel energy amount) and the schedule of the charge / discharge operation may be transmitted to the vehicle.

  Further, the case where the travel energy amount is calculated based on the generated speed pattern has been described as an example. However, the present invention is not limited to this, and the generated speed pattern may be a predetermined speed pattern (for example, a past speed pattern). (Speed pattern in driving) and driving support may be performed based on the comparison result.

  In addition, although the case where three states of no accelerator operation, small accelerator operation, and large accelerator operation are used as the accelerator operation state has been described as an example, the present invention is not limited to this. Four states may be used: no operation, small accelerator operation, during accelerator operation, and large accelerator operation.

  In the above embodiment, the case where the route search device is mounted on the hybrid vehicle driven by both the internal combustion engine and the motor has been described as an example. However, the present invention is not limited to this, and the electric vehicle driven by the motor is not limited thereto. A route search device may be mounted on the automobile.

  Moreover, since the time series data of speed is output and the amount of travel energy is calculated based on it, the total unit of travel energy amount (per unit distance, per unit time, per link, per trip) can be set freely Good.

  Moreover, although the case where the road gradient is acquired from the road network database has been described as an example, the present invention is not limited to this. For example, a front image is captured and the road gradient is estimated from the front image. Good.

Further, the gradient change Δgrad _n may be used as a condition term of the state transition probability, which is a parameter describing the operation characteristics of the individual driver. By setting the state transition probability to Pr (acc _n | grad _n , Δgrad _n ), it is possible to reflect the tendency of the operation at the gradient change point such as from up to down and from down to up.

DESCRIPTION OF SYMBOLS 10,210 Path | route search apparatus 12 Operation amount sensor 13 Vehicle speed sensor 14 Acceleration sensor 16 Position measurement part 20,220 Computer 30 Driving characteristic learning part 32 Operation characteristic memory | storage part 34 Acceleration characteristic memory | storage part 36 Route candidate search part 37 Road gradient acquisition part 38 Operation state prediction unit 40 Speed pattern generation unit 42 Travel energy calculation unit 44 Minimum energy route selection unit 236 Route search unit 244 Charge / discharge schedule determination unit

Claims (9)

Road condition acquisition means for acquiring the road condition at each point of the predicted travel path on which the vehicle travels;
Obtained in advance for each of a plurality of types of road conditions for the driver of the vehicle, the state transition probabilities of the plurality of types of operation states when the driver operates the vehicle, and the road conditions at each point acquired by the road state acquisition means Based on the operation state prediction means for predicting the time series data of the operation state in the travel road,
The relationship between the road state obtained in advance for each of the plurality of types of operation states and the acceleration of the vehicle, the road state of each point acquired by the road state acquisition unit, and the operation predicted by the operation state prediction unit Predicting time series data of acceleration of the vehicle on the road based on time series data of state, and speed of the vehicle on the road based on time series data of acceleration of the vehicle on the road A speed prediction means for predicting time series data of
A speed prediction device including:   The speed prediction apparatus according to claim 1, wherein a relationship between the road state and the acceleration of the vehicle for each of the plurality of types of operation states is obtained in advance for the driver.   3. The speed prediction according to claim 1, further comprising: an energy amount calculation unit that calculates an energy amount consumed when the vehicle travels on the travel path based on the time-series data of the speed predicted by the speed prediction unit. apparatus.   The energy amount calculation means is a time series of the speed predicted when the operation state prediction means repeatedly predicts the time series data of the operation state and the speed prediction means predicts the speed time series data. The speed prediction apparatus according to claim 3, wherein the energy amount is calculated for each piece of data to calculate a range of energy amount consumed when traveling on the traveling road.   The speed prediction apparatus according to claim 1, wherein the road state is a road gradient.   The speed prediction apparatus according to claim 5, wherein the plurality of types of road states are a state in which a road gradient is ascending, a road gradient is flat, and a road gradient is in a descending state.   The speed prediction apparatus according to claim 1, wherein the operation state is an accelerator operation amount.   The speed prediction apparatus according to claim 7, wherein the plurality of types of operation states are a state where there is no accelerator operation and a plurality of types of states where the accelerator operation amount is different. Computer
Road state acquisition means for acquiring a road state at each point of the predicted travel path on which the vehicle travels;
Obtained in advance for each of a plurality of types of road conditions for the driver of the vehicle, the state transition probabilities of the plurality of types of operation states when the driver operates the vehicle, and the road conditions at each point acquired by the road state acquisition means Based on the operation state prediction means for predicting the time-series data of the operation state on the travel road, the relationship between the road state and the acceleration of the vehicle obtained in advance for each of the plurality of types of operation state, Based on the road state of each point acquired by the road state acquisition unit and the time series data of the operation state predicted by the operation state prediction unit, time series data of the acceleration of the vehicle on the travel path is predicted. The speed for predicting the time-series data of the speed of the vehicle on the travel path based on the time-series data of the acceleration of the vehicle on the travel path A program for functioning as a means of prediction.

Images