WO2019167511A1 - Mobile body control device and mobile body control method - Google Patents

Mobile body control device and mobile body control method Download PDF

Info

Publication number
WO2019167511A1
WO2019167511A1 PCT/JP2019/002716 JP2019002716W WO2019167511A1 WO 2019167511 A1 WO2019167511 A1 WO 2019167511A1 JP 2019002716 W JP2019002716 W JP 2019002716W WO 2019167511 A1 WO2019167511 A1 WO 2019167511A1
Authority
WO
WIPO (PCT)
Prior art keywords
movement trajectory
target value
moving body
trajectory
movement
Prior art date
Application number
PCT/JP2019/002716
Other languages
French (fr)
Japanese (ja)
Inventor
之寛 斉藤
遼 高橋
諒 渡辺
Original Assignee
ソニー株式会社
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by ソニー株式会社 filed Critical ソニー株式会社
Publication of WO2019167511A1 publication Critical patent/WO2019167511A1/en

Links

Images

Classifications

    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
    • G05D1/02—Control of position or course in two dimensions

Definitions

  • the present disclosure relates to a moving body control device and a moving body control method.
  • Patent Document 1 describes an autonomous robot moving vehicle that tracks a route including a series of directed straight lines and directed arcs using position feedback and continuous curvature.
  • Patent Document 1 The technique described in the above-mentioned Patent Document 1 is for receiving feedback of self-position and reflecting it in a control system to control an autonomous robot moving vehicle, and relates to control for adjusting the self-position to a given target value.
  • the conventional feedback correction as described in Patent Document 1 relates to feedback control in control, and outputs a control signal from the controller to an actuator such as a motor, and the controller outputs the observed value and the target. It has a loop structure in which a deviation of the value is calculated and a feedback control signal is output to the actuator.
  • a target value for controlling a moving body an action planning unit that plans a movement trajectory of the moving body predicted from the target value, and the moving body is actually operated based on the target value.
  • a control device for a moving body comprising: an actual movement locus acquisition unit that acquires an actual movement locus that has moved; and a movement locus correction unit that corrects the movement locus based on the movement locus and the actual movement locus.
  • the target value for controlling the moving body, the movement locus of the moving body predicted from the target value are planned, and the moving body has actually moved based on the target value.
  • a method for controlling a moving body comprising: obtaining an actual movement locus; and correcting the movement locus based on the movement locus and the actual movement locus.
  • the present disclosure is intended to predict the “target value” and “target value” in local path planning when a mobile body (vehicle, robot, etc.) capable of autonomous movement moves autonomously to a target point.
  • the present invention relates to a technology for generating a pair of “moving trajectories”.
  • the present disclosure relates to a technique for correcting a “movement locus predicted for a target value” in local path planning according to an actual movement locus.
  • disturbances that cannot be taken into account in the motion model and imperfections in the motion model are corrected from the actual movement trajectory, and the discrepancy between the trajectory generation or the trajectory plan and the actual movement trajectory is reduced.
  • FIG. 1 is a schematic diagram illustrating a configuration of a mobile object control system 1000 according to an embodiment of the present disclosure.
  • the mobile body control system 1000 is mounted on a mobile body.
  • the moving body control system 1000 includes an action planning unit 100 and a control unit 200.
  • the action planning unit 100 plans actions to be performed by the moving body according to the surrounding environment and surrounding situation of the moving body and the purpose intended by the moving body, and determines a target value of the moving body.
  • the control unit 200 controls the actuator based on the target value given from the action planning unit 100.
  • the action planning unit 100 receives information on the position of the target point.
  • a target value is determined by the global path planning unit 110 and the local path planning 120.
  • the target value is sent to the control unit 200.
  • the control unit 200 generates a control signal including a pulse signal such as PWM based on the target value sent from the action planning unit 100, and operates the actuator 300 based on the control signal.
  • the sensor 400 observes the movement of the actuator 300 and sends the observation value obtained by the observation to the control unit 200.
  • the control unit 200 performs control (feedback control) so that the observed value matches the target value given from the action planning unit 100.
  • the actuator 300 and the sensor 400 are configured by hardware. Each component other than the actuator 300 and the sensor 400 can be configured by a circuit (hardware) or a central processing unit such as a CPU and a program (software) for causing the central processing unit to function.
  • Global path planning is also called global trajectory generation and global trajectory planning.
  • the global path planning unit 110 generates a global trajectory to the target point without considering the motion model of the moving object.
  • the global path planning unit 110 generates a long-distance trajectory at a lower rate than the local path planning.
  • Global path planning is trajectory generation or trajectory planning that does not consider a motion model, and is similar to trajectory planning used in car navigation systems.
  • a *, RRT (rapidly-exploring random tree), and the like are known.
  • the motion model corresponds to a motion model in which the motion and behavior of the mobile object are described as mathematical expressions.
  • the motion model corresponds to a motion model in which the motion and behavior of the mobile object are described as mathematical expressions.
  • the moving body is an automobile
  • the automobile cannot move directly in the lateral direction, and thus such a motion restriction is defined by the motion model.
  • Local path planning is also referred to as local trajectory generation and local trajectory planning.
  • the local path planning unit 120 plans a trajectory for causing the moving body to follow the trajectory generated by the global path planning unit 110 in consideration of the motion model of the moving body. In addition, the local path planning unit 120 also performs trajectory planning for obstacle avoidance.
  • the local path planning unit 120 performs near-distance trajectory planning while considering the motion model at a higher rate than the global path planning. For example, when the moving body moves from Tokyo to Hokkaido, the trajectory plan by global path planning is updated every hour, for example. On the other hand, the trajectory plan by the local path planning unit 120 is updated, for example, every few seconds or every few milliseconds. Therefore, the local path planning does not generate a trajectory plan for a relatively long section. As a typical local path planning, DWA (Dynamic Window Approach) and the like are known.
  • DWA Dynamic Window Approach
  • a sensor 400 shown in FIG. 1 is a sensor using an internal sensor such as an acceleration sensor, a gyro sensor, or a wheel encoder, a GPS, a magnetic sensor, a radar, a ToF sensor, a camera, Bluetooth (registered trademark), WiFi, or the like. Including external sensors.
  • the mobile object control system 1000 can calculate the self-position of the mobile object.
  • the moving body control system 1000 can obtain a movement locus (actual movement locus) when the moving body actually moves by continuously performing this self-position calculation.
  • the self-position estimation by the inner world sensor may include odometry calculated by the observed value or kinematics of the inertial measurement unit (IMU).
  • FIG. 2 shows a case where the moving body 1 on which the moving body control system 1000 is mounted moves from the start point 20 to the target point 30 while avoiding the obstacle 40 in the passage 12 surrounded by the wall 10.
  • the global path planning unit 110 generates a global path 50 by performing global trajectory generation and global trajectory planning without considering the motion model.
  • the generation of the global path 50 is performed in the same manner as the route set in the car navigation system, for example, in consideration of the positions of the start point 20 and the target point 30, the shape and position of the passage 12, and the like.
  • the global path 50 is generated as a route having the shortest distance from the start point 20 to the target point 30, for example.
  • the local path planning unit 120 performs local short local path trajectory generation and trajectory planning.
  • the local path planning unit 120 sets a target value based on the trajectory plan, and sends the target value to the control unit 200.
  • the target value is basically set along the global path 50 based on the result of determining the surroundings of the moving body 1 such as the obstacle 40 and the motion state of the moving body 1.
  • the control unit 200 sends a control signal based on the target value to the actuator 300, and drives the actuator 300 based on the control signal. Thereby, various operations such as movement of the moving body 1 are performed.
  • the actuator 300 corresponds to a motor that drives the wheels of the moving body 1, a motor that drives the steering, or the like.
  • FIG. 3 is a schematic diagram showing in detail the specific configuration of the local path planning unit 120 and the control unit 200 and the processing performed by the local path planning unit 120.
  • the local path planning unit 120 includes a movement locus calculation unit 122, a movement locus evaluation unit 124, a movement locus selection unit 126, and a movement locus correction unit 128.
  • the control unit 200 includes a control signal generation unit 202 and an actual movement locus acquisition unit.
  • the process of the local path planning unit 120 can be divided into three processes of step S10 to step S30.
  • the processing of each step corresponds to the processing performed by the movement trajectory calculation unit 122, the movement trajectory evaluation unit 124, and the movement trajectory selection unit 126.
  • the control signal generation unit 202 of the control unit 200 generates a control signal for driving the actuator 300 based on the target value sent from the local path planning unit 120.
  • the actual movement trajectory acquisition unit 204 of the control unit 200 acquires the actual movement trajectory of the moving body 1 based on the observation value obtained by the sensor 400 by observation.
  • the actual movement trajectory of the moving body 1 is obtained by the above-described self-position estimation.
  • step S ⁇ b> 10 the movement trajectory calculation unit 122 calculates the movement trajectory of the moving body 1.
  • step S10 a plurality of pairs of “target value” and “movement locus predicted for the target value” are calculated.
  • the “target value” corresponds to the steering angle and the accelerator opening when the moving body 1 is an automobile.
  • the “movement trajectory predicted for the target value” is also different. For example, when the current steering angle is 10 ° to the right and the steering angle range in which the moving body 1 can move by the time of the next control cycle is 20 °, the steering angle of 20 ° is obtained by dividing it into five equal parts.
  • Five target values can be set. In FIG. 3, five different movement loci 60, 62, 64, 66, 68 corresponding to five different target values are shown. In this way, in step S10, a plurality of different movement trajectories corresponding to a plurality of different target values are calculated.
  • the movement trajectory evaluation unit 124 evaluates a plurality of pairs of “target value” and “movement trajectory predicted for the target value” calculated in step S10.
  • the movement trajectory selection unit 126 selects the best one of a plurality of pairs of “target value” and “movement trajectory predicted for the target value”. Then, a target value corresponding to the movement trajectory evaluated as the best is sent to the control unit 200.
  • FIG. 4 is a schematic diagram for explaining the evaluation process in step S20 in the local path planning unit 120.
  • an obstacle 40 exists in the traveling direction of the moving body 1 (the direction along the global path 50).
  • the local path planning it is desirable to set the movement trajectory in the direction along the global path 50 as much as possible.
  • the movement trajectory 64 or the movement trajectory 66 is selected, there is a possibility of colliding with the obstacle 40.
  • the trajectory plan of the local path is greatly deviated from the global path 50.
  • step S20 in FIG. 3 these factors are evaluated, and in the example shown in FIG. 4, the movement locus 62 is the best movement locus. Select a target value.
  • factors such as the deviation from the global path 50 and the possibility of collision with the obstacle 40 are included in the evaluation items when the movement trajectory is evaluated in step S20. Note that the possibility of collision with the obstacle 40 can be determined from the image of the camera included in the sensor 400.
  • the arbitrary factors obtained by observation by the mobile body control system 1000 other than these factors can be included.
  • a plurality of pairs of “target value” and “movement trajectory predicted with respect to the target value” are evaluated, and the target value corresponding to the best movement trajectory is sent to the control unit 200, thereby obstructing the obstacle. It is possible to avoid 40 and move in a direction along the global path 50.
  • FIG. 5A shows a case where the movement trajectory 78 is selected based on the evaluation in step S20 among the plurality of movement trajectories 70, 72, 74, 76, 78 calculated in step S10 of FIG.
  • the target value corresponding to the movement trajectory 78 is sent to the control unit 200, and at the next control cycle, time t + 1, the actuator 300 is used with the target value calculated at time t1.
  • FIG. 5B shows “ideal movement of the moving object” corresponding to the target value at time t + 1.
  • FIG. 5C shows an actual movement locus 80 that is “actual movement of the moving body” at time t + 1 together with a movement locus 78 that is “ideal movement of the moving body”.
  • FIG. 5D feeds back the “movement locus predicted for the target value” in the next control cycle (time t + 2) in accordance with the actual movement locus 80 of the moving body at time t + 1 shown in FIG. 5C.
  • the corrected movement locus 82 is shown.
  • the movement locus 78 is selected by the evaluation in step S20, and the control unit 200 controls the actuator 300 based on the target value corresponding to the movement locus 78.
  • the “ideal movement of the moving body” shown in FIG. 5B and the “actual movement of the moving body” shown in FIG. 5C should match, but in reality, the “ideal movement of the moving body” shown in FIG. In most cases, the “movement” and the “actual movement of the moving object” shown in FIG. 5C do not coincide with each other.
  • the causes include the insufficient construction of the motion model when calculating the movement trajectory and the occurrence of unexpected disturbances.
  • a target value of 40 km / h indicating the vehicle speed is sent to the control unit 200.
  • the car actually travels straight over a distance of 10 m per second.
  • slipping occurs when traveling on a gravel road, and even if the same target value of 40 km / h is sent to the control unit 200, it does not always advance 10 m per second.
  • it is practically difficult to control the moving body 1 in consideration of a perfect motion model and all disturbances.
  • the actual movement locus 80 of the moving body 1 is deviated from the movement locus 78 predicted from the target value.
  • the movement locus predicted from the target value in the next control cycle is feedback-corrected.
  • the target value is calculated based on the actual movement locus 80 shown in FIG. 5C.
  • the predicted movement locus 78 is corrected.
  • the movement locus 78 predicted from the target value is corrected to the movement locus 82.
  • the corrected movement locus 82 matches the actual movement locus 80 in the previous control period (time t + 1) shown in FIG. 5B.
  • the trajectory plan by the local path planning unit 120 is updated, for example, every few seconds or every several milliseconds, and this update cycle becomes the control cycle.
  • the correspondence between the movement trajectory predicted for the target value and the actual movement trajectory can also be determined for a certain period of time. A lot of data will be acquired. Therefore, with respect to the target value in the past control cycle, the actual trajectory for the target value in the past control cycle is used to adaptively move the predicted trajectory from the target value in any subsequent control cycle. It can be corrected.
  • the difference between the movement trajectory predicted from the target value and the actual movement trajectory decreases every time the cycle is repeated, so that the movement trajectory predicted for the target value by data accumulation is adapted to the actual movement trajectory. Is possible.
  • FIG. 6 is a schematic diagram showing processing of the local path planning unit 120 with feedback correction added. Steps S10, S20, and S30 shown in FIG. 6 correspond to steps S10, S20, and S30 in FIG.
  • a process of correcting the movement locus is added to the process of step S10 of FIG.
  • the movement trajectory is corrected by the movement trajectory correction unit 128 shown in FIG.
  • the target trajectory of the moving body 1 and the trajectory predicted for the target value are calculated in step S10, and the trajectory is corrected.
  • Steps S20 and S30 are performed in the same manner as in FIG.
  • FIG. 5C and 5D show an example in which only the movement trajectory 78 is corrected among the plurality of calculated movement trajectories 70, 72, 74, 76, 78.
  • FIG. 5D other movement trajectories 70, 72, 74 and 76 can also be corrected.
  • the movement locus 78 predicted for the target value is determined as described above.
  • the movement locus 78 can be expressed by the speed V of the automobile and the angular speed ⁇ (yaw rate) of the turn. That is, the speed V and the angular speed ⁇ are obtained from the accelerator opening that is the target value and the steering angle, and these represent the movement trajectory. As shown in FIG.
  • ⁇ In the case of 1.5 rad / s, the angular velocity of the actual movement locus 80 is 1.5 times the angular velocity of the movement locus 78.
  • the movement trajectory 78 is corrected based on the actual movement trajectory 80, and other movement trajectories 76 are corrected based on the difference between the movement trajectory 78 and the actual movement trajectory 80.
  • the moving locus 76 is also corrected so that the angular velocity is 1.5 times.
  • the angular velocity of the movement locus 76 is multiplied by 1.5 to correct the movement locus 84.
  • the angular velocity defining the movement trajectory 84 is 0.75 rad / s.
  • more general correction can be performed by applying the parameter indicating the deviation between the predicted movement trajectory for the target value and the actual movement trajectory to the movement trajectories of other target values. Can do.
  • FIG. 8 when the movement locus 78 is corrected by the method of FIGS. 5A to 5D, the movement locus 78 is not corrected to the actual movement locus 80, but the movement locus 78 is moved between the movement locus 78 and the actual movement locus 80.
  • the correction is performed.
  • the corrected movement locus 86 does not have to coincide with the actual movement locus 80.
  • some parameters for calculating the movement locus predicted from the target value are changed, and the next control The movement trajectory predicted from the target value in the cycle is corrected.
  • the present embodiment has been described by taking movement by a moving body 1 such as an automobile as an example.
  • the present disclosure is not limited to application to a moving body such as an automobile, and generates a trajectory in consideration of a motion model It can be widely applied when performing trajectory planning.
  • the present disclosure can be applied to a wide range of uses such as a support arm that supports a medical endoscope, a humanoid, a pet-type robot, and a transfer robot.
  • the deviation between the movement locus predicted with respect to the target value and the actual movement locus is reduced, and the moving object 1
  • the autonomous movement performance of the vehicle can be improved, the tracking performance to the global path can be improved, and the obstacle avoidance performance can be improved. As a result, it is possible to suppress the rattling of the control that occurs during automatic driving.
  • the moving body 1 adapts to the environment so that the difference between the movement trajectory predicted from the target value and the actual movement trajectory is reduced without considering a complicated motion model or disturbance, local path planning is performed.
  • the implementation itself is simplified and the calculation cost can be reduced.
  • the moving body 1 adapts at any time even in an environment where local path planning is not assumed, it is possible to move under various situations.
  • a target value for controlling a moving body an action planning unit that plans a movement trajectory of the moving body predicted from the target value, An actual movement trajectory acquisition unit that acquires an actual movement trajectory in which the moving body has actually moved based on the target value; A movement locus correction unit for correcting the movement locus based on the movement locus and the actual movement locus; A control device for a moving body.
  • the action planning unit plans the movement trajectory predicted from each of the plurality of target values and the plurality of target values, The movement trajectory correction unit corrects the movement trajectory predicted from the specific target value based on the actual movement trajectory corresponding to the specific target value. Control device.
  • the control unit for a moving body according to (2) wherein the movement trajectory correction unit corrects the movement trajectory predicted from the specific target value to the actual movement trajectory.
  • the action planning unit plans the movement trajectory predicted from each of the plurality of target values and the plurality of target values,
  • the moving trajectory correction unit corrects the moving trajectory predicted from the other target values based on the actual moving trajectory corresponding to the specific target value, and controls the moving body according to (1). apparatus.
  • the movement trajectory correction unit corrects the movement trajectory in the current control cycle based on the movement trajectory in the control cycle prior to the current control cycle and the actual movement trajectory, (1) to (4)
  • the control apparatus of the moving body in any one of.
  • the mobile body control device further including a control unit that controls the mobile body based on the target value.
  • the action planning unit includes a global path planning unit that performs a global action plan, and a local path planning unit that performs a local action plan, The target value and the movement trajectory of the mobile body predicted from the target value are planned by the local path planning unit based on the global action plan by the global path planning unit.
  • the control apparatus of the moving body in any one of.

Abstract

[Problem] To resolve a deviation between a movement trajectory forecast with regard to a target value and an actual movement trajectory. [Solution] A mobile body control device according to the present disclosure comprises: an action planning part for planning a target value for controlling a mobile body and a movement trajectory of the mobile body being forecast from the target value; a real movement trajectory acquisition part for acquiring a real movement trajectory whereon the mobile body has actually moved on the basis of the target value; and a movement trajectory correction part for correcting the movement trajectory on the basis of the movement trajectory and the real movement trajectory.

Description

移動体の制御装置及び移動体の制御方法MOBILE BODY CONTROL DEVICE AND MOBILE BODY CONTROL METHOD
　本開示は、移動体の制御装置及び移動体の制御方法に関する。 The present disclosure relates to a moving body control device and a moving body control method.
　従来、例えば下記の特許文献１には、位置のフィードバックと連続曲率を用いて、有向直線と有向円弧の系列からなる経路を追跡する自律ロボット移動車に関して記載されている。 Conventionally, for example, the following Patent Document 1 describes an autonomous robot moving vehicle that tracks a route including a series of directed straight lines and directed arcs using position feedback and continuous curvature.
特開平１１－３２７６４１号公報Japanese Patent Laid-Open No. 11-327641
　上記特許文献１に記載された技術は、自己位置のフィードバックを受けて制御システムに反映し、自律ロボット移動車の制御を行うものであり、与えられた目標値に自己位置を合わせるための制御に関するものである。換言すれば、特許文献１に記載されているような従来のフィードバック補正は、制御におけるフィードバック制御に関するものであり、制御器からモーターなどのアクチュエータへ制御信号を出力し、制御器が観測値と目標値のズレを計算してアクチュエータへフィードバック制御信号を出力するというループ構造になっている。 The technique described in the above-mentioned Patent Document 1 is for receiving feedback of self-position and reflecting it in a control system to control an autonomous robot moving vehicle, and relates to control for adjusting the self-position to a given target value. Is. In other words, the conventional feedback correction as described in Patent Document 1 relates to feedback control in control, and outputs a control signal from the controller to an actuator such as a motor, and the controller outputs the observed value and the target. It has a loop structure in which a deviation of the value is calculated and a feedback control signal is output to the actuator.
　しかしながら、特許文献１に記載されているようなフィードバック制御では、目標値に対してフィードバック補正が行われることがないため、目標値に対して予測される移動軌跡と実際の移動軌跡との間のズレを解消することは困難である。 However, in the feedback control as described in Patent Document 1, feedback correction is not performed on the target value, and therefore, there is no difference between the movement locus predicted for the target value and the actual movement locus. It is difficult to eliminate the deviation.
　そこで、目標値に対して予測される移動軌跡と実際の移動軌跡との間のズレを解消することが求められていた。 Therefore, it has been demanded to eliminate the deviation between the movement locus predicted for the target value and the actual movement locus.
　本開示によれば、移動体を制御するための目標値と、前記目標値から予測される前記移動体の移動軌跡を計画する行動計画部と、前記目標値に基づいて前記移動体が実際に移動した実移動軌跡を取得する実移動軌跡取得部と、前記移動軌跡と前記実移動軌跡に基づいて、前記移動軌跡を補正する移動軌跡補正部と、を備える、移動体の制御装置が提供される。 According to the present disclosure, a target value for controlling a moving body, an action planning unit that plans a movement trajectory of the moving body predicted from the target value, and the moving body is actually operated based on the target value. There is provided a control device for a moving body, comprising: an actual movement locus acquisition unit that acquires an actual movement locus that has moved; and a movement locus correction unit that corrects the movement locus based on the movement locus and the actual movement locus. The
　本開示によれば、移動体を制御するための目標値と、前記目標値から予測される前記移動体の移動軌跡を計画することと、前記目標値に基づいて前記移動体が実際に移動した実移動軌跡を取得することと、前記移動軌跡と前記実移動軌跡に基づいて、前記移動軌跡を補正することと、を備える、移動体の制御方法が提供される。 According to the present disclosure, the target value for controlling the moving body, the movement locus of the moving body predicted from the target value are planned, and the moving body has actually moved based on the target value. There is provided a method for controlling a moving body, comprising: obtaining an actual movement locus; and correcting the movement locus based on the movement locus and the actual movement locus.
　以上説明したように本開示によれば、目標値に対して予測される移動軌跡と実際の移動軌跡との間のズレを解消することが可能となる。
　なお、上記の効果は必ずしも限定的なものではなく、上記の効果とともに、または上記の効果に代えて、本明細書に示されたいずれかの効果、または本明細書から把握され得る他の効果が奏されてもよい。 As described above, according to the present disclosure, it is possible to eliminate the deviation between the movement locus predicted for the target value and the actual movement locus.
Note that the above effects are not necessarily limited, and any of the effects shown in the present specification, or other effects that can be grasped from the present specification, together with or in place of the above effects. May be played.
本開示の一実施形態に係る移動体制御システム１０００の構成を示す模式図である。It is a mimetic diagram showing the composition of mobile control system 1000 concerning one embodiment of this indication. グローバルパスプランニング及びローカルパスプランニングに基づく移動を説明するための模式図である。It is a schematic diagram for demonstrating the movement based on a global path planning and a local path planning. ローカルパスプランニング部で行われる処理を詳細に示す模式図である。It is a schematic diagram which shows the process performed in a local path planning part in detail. ローカルパスプランニング部において、ステップＳ２０の評価の処理を説明するための模式図である。In a local path planning part, it is a schematic diagram for demonstrating the process of evaluation of step S20. ローカルパスプランニングにフィードバック補正をかける方法を説明するための模式図である。It is a schematic diagram for demonstrating the method of applying feedback correction to local path planning. ローカルパスプランニングにフィードバック補正をかける方法を説明するための模式図である。It is a schematic diagram for demonstrating the method of applying feedback correction to local path planning. ローカルパスプランニングにフィードバック補正をかける方法を説明するための模式図である。It is a schematic diagram for demonstrating the method of applying feedback correction to local path planning. ローカルパスプランニングにフィードバック補正をかける方法を説明するための模式図である。It is a schematic diagram for demonstrating the method of applying feedback correction to local path planning. フィードバック補正を加えたローカルパスプランニング部１２０の処理を示す模式図である。It is a schematic diagram which shows the process of the local path planning part 120 which added feedback correction. １の移動軌跡の補正を他の移動軌跡にも反映させる場合を示す模式図である。It is a schematic diagram which shows the case where correction of 1 movement locus | trajectory is reflected also on another movement locus | trajectory. 目標値から予測される移動軌跡と実際の移動軌跡との間の任意の移動軌跡に補正する例を示す模式図である。It is a schematic diagram which shows the example correct | amended to the arbitrary movement locus | trajectory between the movement locus | trajectory estimated from a target value, and an actual movement locus | trajectory.
　以下に添付図面を参照しながら、本開示の好適な実施の形態について詳細に説明する。なお、本明細書及び図面において、実質的に同一の機能構成を有する構成要素については、同一の符号を付することにより重複説明を省略する。 Hereinafter, preferred embodiments of the present disclosure will be described in detail with reference to the accompanying drawings. In addition, in this specification and drawing, about the component which has the substantially same function structure, duplication description is abbreviate | omitted by attaching | subjecting the same code | symbol.
　なお、説明は以下の順序で行うものとする。
　１．本開示の概要
　２．移動体制御システムの構成
　　２．１．グローバルパスプランニング
　　２．２．ローカルパスプランニング
　　２．３．自己位置推定
　３．グローバルパスプランニング及びローカルパスプランニングに基づく移動
　４．ローカルパスプランニングのフィードバック補正 The description will be made in the following order.
1. Overview of this disclosure Configuration of mobile control system 2.1. Global path planning 2.2. Local path planning 2.3. 2. Self-position estimation 3. Travel based on global path planning and local path planning Local path planning feedback correction
　１．本開示の概要
　本開示は、自律移動可能な移動体（車両、ロボット等）が目標となる地点へ自律移動する際に、ローカルパスプランニングにて「目標値」と「目標値に対して予測される移動軌跡」のペアを生成する技術に関する。そして、本開示は、実際の移動軌跡に応じて、ローカルパスプランニングの「目標値に対して予測される移動軌跡」を補正する技術に関する。本開示では、ローカルパスプランニングにおいて、運動モデルで考慮しきれない外乱や運動モデルの不完全性を実際の移動軌跡から補正し、軌道生成または軌道計画と実移動軌跡の乖離を軽減する。 1. Outline of the present disclosure The present disclosure is intended to predict the “target value” and “target value” in local path planning when a mobile body (vehicle, robot, etc.) capable of autonomous movement moves autonomously to a target point. The present invention relates to a technology for generating a pair of “moving trajectories”. The present disclosure relates to a technique for correcting a “movement locus predicted for a target value” in local path planning according to an actual movement locus. In the present disclosure, in local path planning, disturbances that cannot be taken into account in the motion model and imperfections in the motion model are corrected from the actual movement trajectory, and the discrepancy between the trajectory generation or the trajectory plan and the actual movement trajectory is reduced.
　２．移動体制御システムの構成
　図１は、本開示の一実施形態に係る移動体制御システム１０００の構成を示す模式図である。移動体制御システム１０００は、移動体に搭載される。図１に示すように、移動体制御システム１０００は、行動計画部１００と制御部２００を有して構成されている。行動計画部１００は、移動体の周辺環境および周辺の状況と移動体の意図する目的に応じて、移動体が成すべき行動を計画し、移動体の目標値を定める。制御部２００は、行動計画部１００から与えられた目標値に基づいてアクチュエータを制御する。 2. Configuration of Mobile Object Control System FIG. 1 is a schematic diagram illustrating a configuration of a mobile object control system 1000 according to an embodiment of the present disclosure. The mobile body control system 1000 is mounted on a mobile body. As shown in FIG. 1, the moving body control system 1000 includes an action planning unit 100 and a control unit 200. The action planning unit 100 plans actions to be performed by the moving body according to the surrounding environment and surrounding situation of the moving body and the purpose intended by the moving body, and determines a target value of the moving body. The control unit 200 controls the actuator based on the target value given from the action planning unit 100.
　例えば、移動体の目標地点が与えられると、行動計画部１００が目標地点の位置に関する情報を受け取る。行動計画部１００では、グローバルパスプランニング部１１０とローカルパスプランニング１２０により、目標値を定める。目標値は、制御部２００に送られる。 For example, when the target point of the moving object is given, the action planning unit 100 receives information on the position of the target point. In the action planning unit 100, a target value is determined by the global path planning unit 110 and the local path planning 120. The target value is sent to the control unit 200.
　制御部２００は、行動計画部１００から送られた目標値に基づいて、ＰＷＭなどのパルス信号からなる制御信号を生成し、制御信号に基づいてアクチュエータ３００を動作させる。センサ４００は、アクチュエータ３００の動きを観測し、観測により得た観測値を制御部２００に送る。制御部２００は、観測値が行動計画部１００から与えられた目標値と一致するように制御（フィードバック制御）を行う。なお、図１において、アクチュエータ３００とセンサ４００はハードウェアにより構成される。アクチュエータ３００とセンサ４００以外の各構成要素は、回路（ハードウェア）、またはＣＰＵなどの中央演算処理装置とこれを機能させるためのプログラム（ソフトウェア）から構成することができる。 The control unit 200 generates a control signal including a pulse signal such as PWM based on the target value sent from the action planning unit 100, and operates the actuator 300 based on the control signal. The sensor 400 observes the movement of the actuator 300 and sends the observation value obtained by the observation to the control unit 200. The control unit 200 performs control (feedback control) so that the observed value matches the target value given from the action planning unit 100. In FIG. 1, the actuator 300 and the sensor 400 are configured by hardware. Each component other than the actuator 300 and the sensor 400 can be configured by a circuit (hardware) or a central processing unit such as a CPU and a program (software) for causing the central processing unit to function.
　　２．１．グローバルパスプランニング
　グローバルパスプランニングは、大域的軌道生成、大域的軌道計画とも称される。グローバルパスプランニング部１１０では、移動体の運動モデルを考慮せずに、目標地点までの大域的な軌道の生成を行う。グローバルパスプランニング部１１０では、ローカルパスプランニングと比較すると、低レートで長距離の軌道生成を行う。グローバルパスプランニングは、運動モデルを考慮しない軌道生成または軌道計画であり、カーナビゲーションシステムで使用される軌道計画と同様のものである。グローバルパスプランニングの代表的なものとして、Ａ＊、ＲＲＴ（Ｒａｐｉｄｌｙ－ｅｘｐｌｏｒｉｎｇ　Ｒａｎｄｏｍ　Ｔｒｅｅ）などが知られている。 2.1. Global path planning Global path planning is also called global trajectory generation and global trajectory planning. The global path planning unit 110 generates a global trajectory to the target point without considering the motion model of the moving object. The global path planning unit 110 generates a long-distance trajectory at a lower rate than the local path planning. Global path planning is trajectory generation or trajectory planning that does not consider a motion model, and is similar to trajectory planning used in car navigation systems. As typical global path planning, A *, RRT (rapidly-exploring random tree), and the like are known.
　なお、運動モデルとは、移動体の動作、振る舞いを数式として記述したもの等が該当する。例えば、移動体が自動車の場合、自動車は横方向へ直接移動できないため、このような動作の制約は運動モデルによって規定される。 Note that the motion model corresponds to a motion model in which the motion and behavior of the mobile object are described as mathematical expressions. For example, when the moving body is an automobile, the automobile cannot move directly in the lateral direction, and thus such a motion restriction is defined by the motion model.
　　２．２．ローカルパスプランニング
　ローカルパスプランニングは、局所的軌道生成、局所的軌道計画とも称される。ローカルパスプランニング部１２０では、移動体の運動モデルを考慮して、グローバルパスプランニング部１１０で生成された軌道に移動体を追従させるための軌道を計画する。また、ローカルパスプランニング部１２０では、障害物回避の軌道計画も行う。 2.2. Local path planning Local path planning is also referred to as local trajectory generation and local trajectory planning. The local path planning unit 120 plans a trajectory for causing the moving body to follow the trajectory generated by the global path planning unit 110 in consideration of the motion model of the moving body. In addition, the local path planning unit 120 also performs trajectory planning for obstacle avoidance.
　ローカルパスプランニング部１２０では、グローバルパスプランニングと比較して、より高いレートで、運動モデルを考慮しながら近距離の軌道計画を行う。例えば、移動体が東京から北海道へ移動するような場合に、グローバルパスプランニングによる軌道計画は、例えば１時間毎に更新される。これに対し、ローカルパスプランニング部１２０による軌道計画は、例えば数秒毎、数ミリ秒毎に更新される。従って、ローカルパスプランニングでは、比較的長い区間の軌道計画は生成しない。ローカルパスプランニングの代表的なものとして、ＤＷＡ（Ｄｙｎａｍｉｃ　Ｗｉｎｄｏｗ　Ａｐｐｒｏａｃｈ）などが知られている。 The local path planning unit 120 performs near-distance trajectory planning while considering the motion model at a higher rate than the global path planning. For example, when the moving body moves from Tokyo to Hokkaido, the trajectory plan by global path planning is updated every hour, for example. On the other hand, the trajectory plan by the local path planning unit 120 is updated, for example, every few seconds or every few milliseconds. Therefore, the local path planning does not generate a trajectory plan for a relatively long section. As a typical local path planning, DWA (Dynamic Window Approach) and the like are known.
　　２．３．自己位置推定
　図１に示すセンサ４００は、加速度センサー、ジャイロセンサー、車輪エンコーダーなどの内界センサと、ＧＰＳ、磁気センサー、レーダ、ＴｏＦセンサ、カメラ、Ｂｌｕｅｔｏｏｔｈ（登録商標）やＷｉｆｉなどを利用したセンサ等の外界センサを含む。これらのセンサ４００の観測によって、移動体制御システム１０００は、移動体の自己位置を算出することができる。移動体制御システム１０００は、この自己位置算出を連続的に行うことで、移動体が実際に移動した際の移動軌跡（実移動軌跡）を求めることができる。なお、移動体の自己位置を算出する際には、内界センサと外界センサのいずれか一方のみを用いても良いし、双方を用いても良い。内界センサによる自己位置推定は、慣性計測装置（ＩＭＵ）の観測値や運動力学により演算したオドメトリを含むものであっても良い。 2.3. Self-position estimation A sensor 400 shown in FIG. 1 is a sensor using an internal sensor such as an acceleration sensor, a gyro sensor, or a wheel encoder, a GPS, a magnetic sensor, a radar, a ToF sensor, a camera, Bluetooth (registered trademark), WiFi, or the like. Including external sensors. By observing these sensors 400, the mobile object control system 1000 can calculate the self-position of the mobile object. The moving body control system 1000 can obtain a movement locus (actual movement locus) when the moving body actually moves by continuously performing this self-position calculation. In calculating the self position of the moving body, only one of the inner world sensor and the outer world sensor or both of them may be used. The self-position estimation by the inner world sensor may include odometry calculated by the observed value or kinematics of the inertial measurement unit (IMU).
　３．グローバルパスプランニング及びローカルパスプランニングに基づく移動
　次に、図２に基づいて、グローバルパスプランニング及びローカルパスプランニングに基づく移動体１の移動について説明する。図２では、移動体制御システム１０００を搭載する移動体１が、壁１０に囲まれた通路１２内を開始地点２０から目標地点３０まで障害物４０を避けながら移動する場合を示している。 3. Movement Based on Global Path Planning and Local Path Planning Next, movement of the moving body 1 based on global path planning and local path planning will be described based on FIG. FIG. 2 shows a case where the moving body 1 on which the moving body control system 1000 is mounted moves from the start point 20 to the target point 30 while avoiding the obstacle 40 in the passage 12 surrounded by the wall 10.
　まず初めに、グローバルパスプランニング部１１０が運動モデルを考慮しない大域的軌道生成、大域的軌道計画を行い、グローバルパス５０を生成する。グローバルパス５０の生成は、例えばカーナビゲーションシステムで設定される経路と同様に行われ、開始地点２０及び目標地点３０の位置、通路１２の形状、位置等を考慮して行われる。グローバルパス５０は、例えば、開始地点２０から目標地点３０までの距離が最短となる経路として生成される。 First, the global path planning unit 110 generates a global path 50 by performing global trajectory generation and global trajectory planning without considering the motion model. The generation of the global path 50 is performed in the same manner as the route set in the car navigation system, for example, in consideration of the positions of the start point 20 and the target point 30, the shape and position of the passage 12, and the like. The global path 50 is generated as a route having the shortest distance from the start point 20 to the target point 30, for example.
　次に、ローカルパスプランニング部１２０が局所的な短いローカルパスの軌道生成、軌道計画を行う。ローカルパスプランニング部１２０は、軌道計画に基づいて目標値を設定し、目標値を制御部２００へ送る。目標値は、基本的にはグローバルパス５０に沿い、障害物４０など移動体１の周囲の状況、移動体１の運動状態を判断した結果に基づいて設定される。 Next, the local path planning unit 120 performs local short local path trajectory generation and trajectory planning. The local path planning unit 120 sets a target value based on the trajectory plan, and sends the target value to the control unit 200. The target value is basically set along the global path 50 based on the result of determining the surroundings of the moving body 1 such as the obstacle 40 and the motion state of the moving body 1.
　制御部２００は、目標値に基づく制御信号をアクチュエータ３００に送り、制御信号に基づいてアクチュエータ３００を駆動させる。これにより、移動体１の移動などの各種動作が行われる。なお、この場合、アクチュエータ３００は、移動体１の車輪を駆動するモータ、ステアリングを駆動するモータ等に該当する。 The control unit 200 sends a control signal based on the target value to the actuator 300, and drives the actuator 300 based on the control signal. Thereby, various operations such as movement of the moving body 1 are performed. In this case, the actuator 300 corresponds to a motor that drives the wheels of the moving body 1, a motor that drives the steering, or the like.
　図３は、ローカルパスプランニング部１２０と制御部２００の具体的な構成と、ローカルパスプランニング部１２０で行われる処理を詳細に示す模式図である。図３に示すように、ローカルパスプランニング部１２０は、移動軌跡算出部１２２、移動軌跡評価部１２４、移動軌跡選択部１２６、移動軌跡補正部１２８を有して構成されている。また、制御部２００は、制御信号生成部２０２と実移動軌跡取得部を有して構成されている。 FIG. 3 is a schematic diagram showing in detail the specific configuration of the local path planning unit 120 and the control unit 200 and the processing performed by the local path planning unit 120. As shown in FIG. 3, the local path planning unit 120 includes a movement locus calculation unit 122, a movement locus evaluation unit 124, a movement locus selection unit 126, and a movement locus correction unit 128. The control unit 200 includes a control signal generation unit 202 and an actual movement locus acquisition unit.
　図３に示すように、ローカルパスプランニング部１２０の処理は、ステップＳ１０～ステップＳ３０の３つの処理に分けることができる。各ステップの処理は、移動軌跡算出部１２２、移動軌跡評価部１２４、及び移動軌跡選択部１２６で行われる処理に対応する。 As shown in FIG. 3, the process of the local path planning unit 120 can be divided into three processes of step S10 to step S30. The processing of each step corresponds to the processing performed by the movement trajectory calculation unit 122, the movement trajectory evaluation unit 124, and the movement trajectory selection unit 126.
　制御部２００の制御信号生成部２０２は、ローカルパスプランニング部１２０から送られた目標値に基づいて、アクチュエータ３００を駆動するための制御信号を生成する。また、制御部２００の実移動軌跡取得部２０４は、センサ４００が観測により得た観測値に基づいて移動体１の実移動軌跡を取得する。移動体１の実移動軌跡は、上述した自己位置推定によって求まる。 The control signal generation unit 202 of the control unit 200 generates a control signal for driving the actuator 300 based on the target value sent from the local path planning unit 120. In addition, the actual movement trajectory acquisition unit 204 of the control unit 200 acquires the actual movement trajectory of the moving body 1 based on the observation value obtained by the sensor 400 by observation. The actual movement trajectory of the moving body 1 is obtained by the above-described self-position estimation.
　以下では、図３に示すステップＳ１０，Ｓ２０，Ｓ３０の処理に基づいて、ローカルパスプランニング部１２０で行われる処理を説明する。先ず、ステップＳ１０では、移動軌跡算出部１２２が、移動体１の移動軌跡を算出する。 Hereinafter, processing performed by the local path planning unit 120 will be described based on the processing of steps S10, S20, and S30 shown in FIG. First, in step S <b> 10, the movement trajectory calculation unit 122 calculates the movement trajectory of the moving body 1.
　具体的に、ステップＳ１０では、「目標値」と「目標値に対して予測される移動軌跡」のペアを複数算出する。「目標値」とは、一例として、移動体１が自動車の場合は、ステアリングの角度とアクセル開度に該当する。異なる複数の目標値に対応して、「目標値に対して予測される移動軌跡」も異なるものになる。例えば、現在のステアリング角度が右に１０°であり、次の制御周期の時刻までに移動体１が移動できるステアリング角度の範囲２０°である場合、２０°のステアリング角度を５等分して得られる５つの目標値を設定できる。図３中には、５つの異なる目標値に対応する、５つの異なる移動軌跡６０，６２，６４，６６，６８が示されている。このように、ステップＳ１０では、複数の異なる目標値に対応する、複数の異なる移動軌跡を算出する。 Specifically, in step S10, a plurality of pairs of “target value” and “movement locus predicted for the target value” are calculated. As an example, the “target value” corresponds to the steering angle and the accelerator opening when the moving body 1 is an automobile. Corresponding to a plurality of different target values, the “movement trajectory predicted for the target value” is also different. For example, when the current steering angle is 10 ° to the right and the steering angle range in which the moving body 1 can move by the time of the next control cycle is 20 °, the steering angle of 20 ° is obtained by dividing it into five equal parts. Five target values can be set. In FIG. 3, five different movement loci 60, 62, 64, 66, 68 corresponding to five different target values are shown. In this way, in step S10, a plurality of different movement trajectories corresponding to a plurality of different target values are calculated.
　次のステップＳ２０では、移動軌跡評価部１２４が、ステップＳ１０で算出した「目標値」と「目標値に対して予測される移動軌跡」の複数のペアを評価する。次のステップＳ３０では、移動軌跡選択部１２６が、「目標値」と「目標値に対して予測される移動軌跡」の複数のペアのうち、最も良いものを選択する。そして、最も良いと評価された移動軌跡に対応する目標値が制御部２００に送られる。 In the next step S20, the movement trajectory evaluation unit 124 evaluates a plurality of pairs of “target value” and “movement trajectory predicted for the target value” calculated in step S10. In the next step S30, the movement trajectory selection unit 126 selects the best one of a plurality of pairs of “target value” and “movement trajectory predicted for the target value”. Then, a target value corresponding to the movement trajectory evaluated as the best is sent to the control unit 200.
　図４は、ローカルパスプランニング部１２０において、ステップＳ２０の評価の処理を説明するための模式図である。図４に示すように、移動体１の進行方向（グローバルパス５０に沿った方向）には、障害物４０が存在する。ローカルパスプランニングにおいては、できるだけグローバルパス５０に沿った方向に移動軌跡を設定することが望ましいが、移動軌跡６４または移動軌跡６６を選択した場合は障害物４０に衝突する可能性がある。一方、移動軌跡６０または移動軌跡６８を選択した場合は、ローカルパスの軌道計画がグローバルパス５０から大きく外れてしまうことになる。 FIG. 4 is a schematic diagram for explaining the evaluation process in step S20 in the local path planning unit 120. As shown in FIG. 4, an obstacle 40 exists in the traveling direction of the moving body 1 (the direction along the global path 50). In the local path planning, it is desirable to set the movement trajectory in the direction along the global path 50 as much as possible. However, when the movement trajectory 64 or the movement trajectory 66 is selected, there is a possibility of colliding with the obstacle 40. On the other hand, when the movement trajectory 60 or the movement trajectory 68 is selected, the trajectory plan of the local path is greatly deviated from the global path 50.
　図３のステップＳ２０の処理では、これらの要因を評価し、図４に示す例では、移動軌跡６２が最も良い移動軌跡であることから、ステップＳ３０において、移動軌跡６２と移動軌跡６２に対応する目標値を選択する。このように、グローバルパス５０からの乖離、障害物４０との衝突可能性、等の要因が、ステップＳ２０において移動軌跡を評価する際の評価項目に含まれる。なお、障害物４０との衝突可能性は、センサ４００に含まれるカメラの画像から判断できる。また、移動軌跡を評価する際には、これらの要因の他にも、移動体制御システム１０００が観測して得られる任意の要因を含めることができる。 In the process of step S20 in FIG. 3, these factors are evaluated, and in the example shown in FIG. 4, the movement locus 62 is the best movement locus. Select a target value. As described above, factors such as the deviation from the global path 50 and the possibility of collision with the obstacle 40 are included in the evaluation items when the movement trajectory is evaluated in step S20. Note that the possibility of collision with the obstacle 40 can be determined from the image of the camera included in the sensor 400. Moreover, when evaluating a movement locus | trajectory, the arbitrary factors obtained by observation by the mobile body control system 1000 other than these factors can be included.
　以上のように、「目標値」と「目標値に対して予測される移動軌跡」の複数のペアを評価し、最良の移動軌跡に対応する目標値を制御部２００に送ることで、障害物４０を回避するとともに、グローバルパス５０に沿った方向に移動することが可能になる。 As described above, a plurality of pairs of “target value” and “movement trajectory predicted with respect to the target value” are evaluated, and the target value corresponding to the best movement trajectory is sent to the control unit 200, thereby obstructing the obstacle. It is possible to avoid 40 and move in a direction along the global path 50.
　４．ローカルパスプランニングのフィードバック補正
　次に、図５Ａ～図５Ｄに基づいて、ローカルパスプランニングにフィードバック補正をかける方法について説明する。制御部２００が目標値に基づいてアクチュエータ３００を制御した場合に、アクチュエータ３００の駆動によって移動体１が実際に移動する軌跡と、目標値に対して予測された移動軌跡との間にズレがないことが望ましい。なぜなら、ローカルパスプランニングは、ある目標値を制御部２００に送った場合に、目標値に対して予測される移動軌跡（図４の例では、移動軌跡６２）と実際の移動軌跡にズレがなく一致することを前提としているためである。 4). Next, a method for applying feedback correction to local path planning will be described with reference to FIGS. 5A to 5D. When the control unit 200 controls the actuator 300 based on the target value, there is no deviation between the trajectory in which the moving body 1 actually moves by driving the actuator 300 and the movement trajectory predicted for the target value. It is desirable. This is because in local path planning, when a certain target value is sent to the control unit 200, there is no deviation between the movement trajectory predicted for the target value (the movement trajectory 62 in the example of FIG. 4) and the actual movement trajectory. This is because it is assumed that they match.
　図５Ａは、図３のステップＳ１０で算出した複数の移動軌跡７０，７２，７４，７６，７８のうち、ステップＳ２０の評価に基づいて移動軌跡７８を選択した場合を示している。時刻ｔで移動軌跡７８が算出されると、移動軌跡７８に対応する目標値が制御部２００に送られて、次の制御周期である時刻ｔ＋１では、時刻ｔ１で算出された目標値でアクチュエータ３００が制御される。図５Ｂは、時刻ｔ＋１における、目標値に対応する「移動体の理想的な動き」を示している。また、図５Ｃは、時刻ｔ＋１における、「移動体の実際の動き」である実移動軌跡８０を、「移動体の理想的な動き」である移動軌跡７８とともに示している。また、図５Ｄは、図５Ｃに示す時刻ｔ＋１での移動体の実際の移動軌跡８０に応じて、次の制御周期（時刻ｔ＋２）での「目標値に対して予測される移動軌跡」をフィードバック補正した移動軌跡８２を示している。 FIG. 5A shows a case where the movement trajectory 78 is selected based on the evaluation in step S20 among the plurality of movement trajectories 70, 72, 74, 76, 78 calculated in step S10 of FIG. When the movement trajectory 78 is calculated at time t, the target value corresponding to the movement trajectory 78 is sent to the control unit 200, and at the next control cycle, time t + 1, the actuator 300 is used with the target value calculated at time t1. Is controlled. FIG. 5B shows “ideal movement of the moving object” corresponding to the target value at time t + 1. FIG. 5C shows an actual movement locus 80 that is “actual movement of the moving body” at time t + 1 together with a movement locus 78 that is “ideal movement of the moving body”. Further, FIG. 5D feeds back the “movement locus predicted for the target value” in the next control cycle (time t + 2) in accordance with the actual movement locus 80 of the moving body at time t + 1 shown in FIG. 5C. The corrected movement locus 82 is shown.
　図５Ａに示すように、ステップＳ２０の評価により移動軌跡７８が選択され、移動軌跡７８に対応する目標値に基づいて、制御部２００がアクチュエータ３００を制御する。このとき、図５Ｂに示す「移動体の理想的な動き」と図５Ｃに示す「移動体の実際の動き」は一致するべきであるが、実際には、図５Ｂに示す「移動体の理想的な動き」と図５Ｃに示す「移動体の実際の動き」は一致しないことが殆どであり、両者の間にはズレが生じる。原因としては、移動軌跡を算出する際の運動モデルの構築が不十分であること、想定外の外乱が発生すること等が挙げられる。 As shown in FIG. 5A, the movement locus 78 is selected by the evaluation in step S20, and the control unit 200 controls the actuator 300 based on the target value corresponding to the movement locus 78. At this time, the “ideal movement of the moving body” shown in FIG. 5B and the “actual movement of the moving body” shown in FIG. 5C should match, but in reality, the “ideal movement of the moving body” shown in FIG. In most cases, the “movement” and the “actual movement of the moving object” shown in FIG. 5C do not coincide with each other. The causes include the insufficient construction of the motion model when calculating the movement trajectory and the occurrence of unexpected disturbances.
　例えば、移動体１が自動車の場合に、車速を示す時速４０ｋｍの目標値を制御部２００に送るものとする。そして、アスファルトの道路を走行する場合は、自動車が１秒間で１０ｍの距離を真っ直ぐ実際に進むものとする。しかし、砂利道を走行する場合はスリップが発生し、同じ時速４０ｋｍの目標値を制御部２００に送った場合でも、１秒間に１０ｍ進むとは限らない。このように、完璧な運動モデルや全ての外乱を考慮して移動体１を制御することは、現実的には困難が伴う。 For example, when the moving body 1 is an automobile, a target value of 40 km / h indicating the vehicle speed is sent to the control unit 200. And when driving on an asphalt road, it is assumed that the car actually travels straight over a distance of 10 m per second. However, slipping occurs when traveling on a gravel road, and even if the same target value of 40 km / h is sent to the control unit 200, it does not always advance 10 m per second. Thus, it is practically difficult to control the moving body 1 in consideration of a perfect motion model and all disturbances.
　図５Ｃに示す例においても、移動体１の実際の移動軌跡８０は、目標値から予測される移動軌跡７８に対してズレが生じている。 Also in the example shown in FIG. 5C, the actual movement locus 80 of the moving body 1 is deviated from the movement locus 78 predicted from the target value.
　そこで、図５Ｄに示すように、図５Ｃに示す移動体１の実際の移動軌跡８０に合わせて、次の制御周期での目標値から予測される移動軌跡をフィードバック補正する。図５Ｄでは、ステップＳ２０の評価の結果、前制御周期（時刻ｔ＋１）と同じ目標値又は近似する目標値が設定された場合に、図５Ｃに示す実際の移動軌跡８０に基づいて、目標値から予測される移動軌跡７８を補正している。具体的に、図５Ｄでは、図５Ｂと同じ目標値が設定された場合に、目標値から予測される移動軌跡７８を移動軌跡８２に補正している。そして、補正後の移動軌跡８２は、図５Ｂに示す前制御周期（時刻ｔ＋１）における実際の移動軌跡８０と一致している。 Therefore, as shown in FIG. 5D, in accordance with the actual movement locus 80 of the moving body 1 shown in FIG. 5C, the movement locus predicted from the target value in the next control cycle is feedback-corrected. In FIG. 5D, when the same target value as that of the previous control cycle (time t + 1) or a target value to be approximated is set as a result of the evaluation in step S20, the target value is calculated based on the actual movement locus 80 shown in FIG. 5C. The predicted movement locus 78 is corrected. Specifically, in FIG. 5D, when the same target value as that in FIG. 5B is set, the movement locus 78 predicted from the target value is corrected to the movement locus 82. Then, the corrected movement locus 82 matches the actual movement locus 80 in the previous control period (time t + 1) shown in FIG. 5B.
　これにより、図５Ｄに示すように、次の制御周期（時刻ｔ＋２）において前制御周期（時刻ｔ＋１）と同じ目標値又は近似する目標値が設定されると、次の制御周期（時刻ｔ＋２）では、前記制御周期（時刻ｔ＋１）で目標値から予測される移動軌跡７８に対して、移動軌跡７８が移動軌跡８２に補正される。移動軌跡８２は前制御周期（時刻ｔ＋１）の実際の移動軌跡８０と同じであるため、次の制御周期（時刻ｔ＋２）における実際の移動軌跡を、補正した移動軌跡８２により近づけることが可能となる。このように、実際の移動軌跡に応じて目標値に対して予測される移動軌跡を補正、更新することで、目標値に対する移動軌跡のズレを抑制することができる。 As a result, as shown in FIG. 5D, when the same target value as that of the previous control period (time t + 1) or a target value that is approximated is set in the next control period (time t + 2), in the next control period (time t + 2). The movement locus 78 is corrected to the movement locus 82 with respect to the movement locus 78 predicted from the target value in the control cycle (time t + 1). Since the movement locus 82 is the same as the actual movement locus 80 in the previous control period (time t + 1), the actual movement locus in the next control period (time t + 2) can be brought closer to the corrected movement locus 82. . In this way, by correcting and updating the movement trajectory predicted for the target value in accordance with the actual movement trajectory, the shift of the movement trajectory with respect to the target value can be suppressed.
　上述したように、ローカルパスプランニング部１２０による軌道計画は、例えば数秒毎、数ミリ秒毎に更新され、この更新の周期が制御周期となる。短時間の制御周期で目標値と目標値に対して予測される移動軌跡を算出していくことで、目標値に対して予測される移動軌跡と実際の移動軌跡の対応関係についても、一定時間内に数多くのデータが取得されることになる。従って、過去の制御周期での目標値に対して、過去の制御周期での目標値に対する実際の移動軌跡を用いて、その後の任意の制御周期の目標値から予測される移動軌跡を適応的に補正することができる。また、周期を重ねる毎に目標値から予測される移動軌跡と実移動軌跡との乖離が減少するための、データの蓄積により目標値に対して予測される移動軌跡を実移動軌跡に適合することが可能となる。 As described above, the trajectory plan by the local path planning unit 120 is updated, for example, every few seconds or every several milliseconds, and this update cycle becomes the control cycle. By calculating the target value and the movement trajectory predicted for the target value in a short control cycle, the correspondence between the movement trajectory predicted for the target value and the actual movement trajectory can also be determined for a certain period of time. A lot of data will be acquired. Therefore, with respect to the target value in the past control cycle, the actual trajectory for the target value in the past control cycle is used to adaptively move the predicted trajectory from the target value in any subsequent control cycle. It can be corrected. In addition, the difference between the movement trajectory predicted from the target value and the actual movement trajectory decreases every time the cycle is repeated, so that the movement trajectory predicted for the target value by data accumulation is adapted to the actual movement trajectory. Is possible.
　図６は、フィードバック補正を加えたローカルパスプランニング部１２０の処理を示す模式図である。図６に示すステップＳ１０，Ｓ２０，Ｓ３０は、図３のステップＳ１０，Ｓ２０，Ｓ３０に対応している。図６に示すステップＳ１０の処理では、図３のステップＳ１０の処理に対して、移動軌跡を補正する処理が加えられている。移動軌跡の補正は、図３に示す移動軌跡補正部１２８によって行われる。これにより、ステップＳ１０で移動体１の目標値と目標値に対して予測される移動軌跡を算出する際に、移動軌跡が補正される。ステップＳ２０、ステップＳ３０の処理は、図３と同様に行われる。 FIG. 6 is a schematic diagram showing processing of the local path planning unit 120 with feedback correction added. Steps S10, S20, and S30 shown in FIG. 6 correspond to steps S10, S20, and S30 in FIG. In the process of step S10 shown in FIG. 6, a process of correcting the movement locus is added to the process of step S10 of FIG. The movement trajectory is corrected by the movement trajectory correction unit 128 shown in FIG. As a result, the target trajectory of the moving body 1 and the trajectory predicted for the target value are calculated in step S10, and the trajectory is corrected. Steps S20 and S30 are performed in the same manner as in FIG.
　次に、図７に基づいて、１の移動軌跡の補正を他の移動軌跡にも反映させる場合について説明する。図５Ｃ及び図５Ｄでは、算出した複数の移動軌跡７０，７２，７４，７６，７８のうち、移動軌跡７８のみを補正する例を示したが、図５Ｄにおいて、他の移動軌跡７０，７２，７４，７６についても補正を行うことができる。 Next, the case where the correction of one movement locus is reflected on another movement locus will be described with reference to FIG. 5C and 5D show an example in which only the movement trajectory 78 is corrected among the plurality of calculated movement trajectories 70, 72, 74, 76, 78. However, in FIG. 5D, other movement trajectories 70, 72, 74 and 76 can also be corrected.
　移動体１が自動車の場合、上述のように目標値（アクセル開度、ステアリング角度）に対して予測される移動軌跡７８が定まる。そして、移動軌跡７８は、自動車の速度Ｖと旋回の角速度ω（ヨーレート）によって表すことができる。つまり、目標値であるアクセル開度とステアリング角度から速度Ｖと角速度ωが求まり、これらが移動軌跡を表すものとなる。図７に示すように、例えば、目標値に対して予測される移動軌跡７８がＶ＝２０ｋｍ／ｓ、ω＝１ｒａｄ／ｓで表され、実際の移動軌跡８０がＶ＝２０ｋｍ／ｓ、ω＝１．５ｒａｄ／ｓで表される場合、移動軌跡７８の角速度に対して、実際の移動軌跡８０の角速度は１．５倍となっている。 When the moving body 1 is an automobile, the movement locus 78 predicted for the target value (accelerator opening degree, steering angle) is determined as described above. The movement locus 78 can be expressed by the speed V of the automobile and the angular speed ω (yaw rate) of the turn. That is, the speed V and the angular speed ω are obtained from the accelerator opening that is the target value and the steering angle, and these represent the movement trajectory. As shown in FIG. 7, for example, the movement trajectory 78 predicted for the target value is represented by V = 20 km / s and ω = 1 rad / s, and the actual movement trajectory 80 is V = 20 km / s, ω = In the case of 1.5 rad / s, the angular velocity of the actual movement locus 80 is 1.5 times the angular velocity of the movement locus 78.
　このため、次の制御周期においては、移動軌跡７８を実際の移動軌跡８０に基づいて補正するとともに、他の移動軌跡７６についても、移動軌跡７８と実際の移動軌跡８０との乖離に基づいて補正する。上記の例では、移動軌跡７８の角速度に対して、実際の移動軌跡８０の角速度は１．５倍であることから、移動軌跡７６についても、角速度が１．５倍となるように補正する。例えば、移動軌跡７６がＶ＝２０ｋｍ／ｓ、ω＝０．５ｒａｄ／ｓで表される場合、移動軌跡７６の角速度を１．５倍して、移動軌跡８４に補正する。これにより、移動軌跡８４を規定する角速度は０．７５ｒａｄ／ｓとなる。 Therefore, in the next control cycle, the movement trajectory 78 is corrected based on the actual movement trajectory 80, and other movement trajectories 76 are corrected based on the difference between the movement trajectory 78 and the actual movement trajectory 80. To do. In the above example, since the actual angular velocity of the moving locus 80 is 1.5 times the angular velocity of the moving locus 78, the moving locus 76 is also corrected so that the angular velocity is 1.5 times. For example, when the movement locus 76 is represented by V = 20 km / s and ω = 0.5 rad / s, the angular velocity of the movement locus 76 is multiplied by 1.5 to correct the movement locus 84. As a result, the angular velocity defining the movement trajectory 84 is 0.75 rad / s.
　以上のように、目標値に対して予測される移動軌跡と実際の移動軌跡との乖離を示すパラメータを、他の目標値の移動軌跡にも適用することで、より汎用的な補正を行うことができる。 As described above, more general correction can be performed by applying the parameter indicating the deviation between the predicted movement trajectory for the target value and the actual movement trajectory to the movement trajectories of other target values. Can do.
　図８は、図５Ａ～図５Ｄの手法で移動軌跡７８を補正する場合に、移動軌跡７８を実移動軌跡８０まで補正するのではなく、移動軌跡７８を移動軌跡７８と実動軌跡８０の間の任意の移動軌跡８６に補正する例を示す模式図である。例えば、移動軌跡７８がＶ＝２０ｋｍ／ｓ、角速度が１．０ｒａｄ／ｓで表され、実際の移動軌跡８０がＶ＝２０ｋｍ／ｓ、ω＝１．５ｒａｄ／ｓで表される場合、補正後の移動軌跡８６の角速度がω＝１．２５ｒａｄ／ｓとなるように補正を行う。このように、補正後の移動軌跡８６を実移動軌跡８０と一致させなくても良い。以上のように、本実施形態では、目標値から予測される移動軌跡と実移動軌跡との乖離に基づいて、目標値から予測される移動軌跡を算出するための何らかのパラメータを変更し、次制御周期で目標値から予測される移動軌跡を補正するようにする。 In FIG. 8, when the movement locus 78 is corrected by the method of FIGS. 5A to 5D, the movement locus 78 is not corrected to the actual movement locus 80, but the movement locus 78 is moved between the movement locus 78 and the actual movement locus 80. It is a schematic diagram which shows the example correct | amended to arbitrary movement locus | trajectory 86 of. For example, when the movement locus 78 is represented by V = 20 km / s and the angular velocity is represented by 1.0 rad / s, and the actual movement locus 80 is represented by V = 20 km / s and ω = 1.5 rad / s, the correction is performed. Is corrected so that the angular velocity of the movement locus 86 becomes ω = 1.25 rad / s. In this way, the corrected movement locus 86 does not have to coincide with the actual movement locus 80. As described above, in the present embodiment, based on the difference between the movement locus predicted from the target value and the actual movement locus, some parameters for calculating the movement locus predicted from the target value are changed, and the next control The movement trajectory predicted from the target value in the cycle is corrected.
　以上の説明では、自動車のような移動体１による移動を例として本実施形態を説明してが、本開示は自動車のような移動体への適用に限らず、運動モデルを考慮して軌道生成、軌道計画を行う場合に広く適用ですることができる。本開示は、例えば医療用の内視鏡などを支持する支持アーム、ヒューマノイド、ペット型ロボット、搬送用ロボットなど幅広い用途への適用が可能である。 In the above description, the present embodiment has been described by taking movement by a moving body 1 such as an automobile as an example. However, the present disclosure is not limited to application to a moving body such as an automobile, and generates a trajectory in consideration of a motion model It can be widely applied when performing trajectory planning. The present disclosure can be applied to a wide range of uses such as a support arm that supports a medical endoscope, a humanoid, a pet-type robot, and a transfer robot.
　以上説明したように本実施形態によれば、ローカルパスプランニングにフィードバック補正をかけることで、目標値に対して予測される移動軌跡と実際の移動軌跡との間のズレが減少し、移動体１の自律移動性能が向上し、グローバルパスへの追従性能を高めるとともに障害物回避性能を向上することができる。これにより、自動運転の際に発生する制御のガタツキを抑制することも可能となる。 As described above, according to the present embodiment, by applying feedback correction to local path planning, the deviation between the movement locus predicted with respect to the target value and the actual movement locus is reduced, and the moving object 1 The autonomous movement performance of the vehicle can be improved, the tracking performance to the global path can be improved, and the obstacle avoidance performance can be improved. As a result, it is possible to suppress the rattling of the control that occurs during automatic driving.
　この結果、ローカルパスプランニングのフィードバック補正がない場合に比べ、移動時の制御のガタツキが減ったり、避けられなかった障害物を回避できるようになる。特に人命の関わる車の自動運転などでは重要な技術となる。 As a result, compared to the case where there is no feedback correction for local path planning, the control rattling during movement is reduced and obstacles that could not be avoided can be avoided. This is especially important for automated driving of vehicles that involve human lives.
　また、複雑な運動モデルや外乱を考慮することなく、目標値から予測される移動軌跡と実際の移動軌跡との差が少なくなるように移動体１が環境に適応していくため、ローカルパスプランニング自体の実装が簡単になり、且つ、計算コストを削減することも可能となる。これにより、複雑な運動モデルを数式等を用いて構築することが不要となる。従って、例えば性能の低い廉価なコンピュータで移動体制御システム１０００を構築することも可能であり、製造コストの削減を達成できる。また、開発時間の削減を達成することも可能となる。更に、ローカルパスプランニングが想定されない環境でも移動体１が随時適応していくため、さまざまな状況下での移動が可能となる。 In addition, since the moving body 1 adapts to the environment so that the difference between the movement trajectory predicted from the target value and the actual movement trajectory is reduced without considering a complicated motion model or disturbance, local path planning is performed. The implementation itself is simplified and the calculation cost can be reduced. This eliminates the need to construct a complex motion model using mathematical formulas or the like. Therefore, for example, it is possible to construct the mobile control system 1000 with an inexpensive computer with low performance, and a reduction in manufacturing cost can be achieved. It is also possible to achieve a reduction in development time. Furthermore, since the moving body 1 adapts at any time even in an environment where local path planning is not assumed, it is possible to move under various situations.
　以上、添付図面を参照しながら本開示の好適な実施形態について詳細に説明したが、本開示の技術的範囲はかかる例に限定されない。本開示の技術分野における通常の知識を有する者であれば、特許請求の範囲に記載された技術的思想の範疇内において、各種の変更例または修正例に想到し得ることは明らかであり、これらについても、当然に本開示の技術的範囲に属するものと了解される。 The preferred embodiments of the present disclosure have been described in detail above with reference to the accompanying drawings, but the technical scope of the present disclosure is not limited to such examples. It is obvious that a person having ordinary knowledge in the technical field of the present disclosure can come up with various changes or modifications within the scope of the technical idea described in the claims. Of course, it is understood that it belongs to the technical scope of the present disclosure.
　また、本明細書に記載された効果は、あくまで説明的または例示的なものであって限定的ではない。つまり、本開示に係る技術は、上記の効果とともに、または上記の効果に代えて、本明細書の記載から当業者には明らかな他の効果を奏しうる。 In addition, the effects described in this specification are merely illustrative or illustrative, and are not limited. That is, the technology according to the present disclosure can exhibit other effects that are apparent to those skilled in the art from the description of the present specification in addition to or instead of the above effects.
　なお、以下のような構成も本開示の技術的範囲に属する。
（１）　移動体を制御するための目標値と、前記目標値から予測される前記移動体の移動軌跡を計画する行動計画部と、
　前記目標値に基づいて前記移動体が実際に移動した実移動軌跡を取得する実移動軌跡取得部と、
　前記移動軌跡と前記実移動軌跡に基づいて、前記移動軌跡を補正する移動軌跡補正部と、
　を備える、移動体の制御装置。
（２）　前記行動計画部は、複数の前記目標値と複数の前記目標値のそれぞれから予測される前記移動軌跡を計画し、
　前記移動軌跡補正部は、特定の前記目標値に対応する前記実移動軌跡に基づいて、前記特定の前記目標値から予測される前記移動軌跡を補正する、前記（１）に記載の移動体の制御装置。
（３）　前記移動軌跡補正部は、前記特定の前記目標値から予測される前記移動軌跡を前記実移動軌跡に補正する、前記（２）に記載の移動体の制御装置。
（４）　前記行動計画部は、複数の前記目標値と複数の前記目標値のそれぞれから予測される前記移動軌跡を計画し、
　前記移動軌跡補正部は、特定の前記目標値に対応する前記実移動軌跡に基づいて、他の前記目標値から予測される前記移動軌跡を補正する、前記（１）に記載の移動体の制御装置。
（５）　前記移動軌跡補正部は、現制御周期より以前の制御周期における前記移動軌跡と前記実移動軌跡に基づいて、現制御周期における前記移動軌跡を補正する、前記（１）～（４）のいずれかに記載の移動体の制御装置。
（６）　前記目標値に基づいて前記移動体を制御する制御部を備える、前記（１）～（５）のいずれかに記載の移動体の制御装置。
（７）　前記行動計画部は、大域的な行動計画を行うグローバルパスプランニング部と、局所的な行動計画を行うローカルパスプランニング部とを含み、
　前記目標値と、前記目標値から予測される前記移動体の移動軌跡は、前記グローバルパスプランニング部による前記大域的な行動計画に基づいて、前記ローカルパスプランニング部が計画する、前記（１）～（６）のいずれかに記載の移動体の制御装置。
（８）　移動体を制御するための目標値と、前記目標値から予測される前記移動体の移動軌跡を計画することと、
　前記目標値に基づいて前記移動体が実際に移動した実移動軌跡を取得することと、
　前記移動軌跡と前記実移動軌跡に基づいて、前記移動軌跡を補正することと、
　を備える、移動体の制御方法 The following configurations also belong to the technical scope of the present disclosure.
(1) A target value for controlling a moving body, an action planning unit that plans a movement trajectory of the moving body predicted from the target value,
An actual movement trajectory acquisition unit that acquires an actual movement trajectory in which the moving body has actually moved based on the target value;
A movement locus correction unit for correcting the movement locus based on the movement locus and the actual movement locus;
A control device for a moving body.
(2) The action planning unit plans the movement trajectory predicted from each of the plurality of target values and the plurality of target values,
The movement trajectory correction unit corrects the movement trajectory predicted from the specific target value based on the actual movement trajectory corresponding to the specific target value. Control device.
(3) The control unit for a moving body according to (2), wherein the movement trajectory correction unit corrects the movement trajectory predicted from the specific target value to the actual movement trajectory.
(4) The action planning unit plans the movement trajectory predicted from each of the plurality of target values and the plurality of target values,
The moving trajectory correction unit corrects the moving trajectory predicted from the other target values based on the actual moving trajectory corresponding to the specific target value, and controls the moving body according to (1). apparatus.
(5) The movement trajectory correction unit corrects the movement trajectory in the current control cycle based on the movement trajectory in the control cycle prior to the current control cycle and the actual movement trajectory, (1) to (4) The control apparatus of the moving body in any one of.
(6) The mobile body control device according to any one of (1) to (5), further including a control unit that controls the mobile body based on the target value.
(7) The action planning unit includes a global path planning unit that performs a global action plan, and a local path planning unit that performs a local action plan,
The target value and the movement trajectory of the mobile body predicted from the target value are planned by the local path planning unit based on the global action plan by the global path planning unit. (6) The control apparatus of the moving body in any one of.
(8) Planning a target value for controlling the moving body, and a movement locus of the moving body predicted from the target value;
Acquiring an actual movement trajectory in which the moving body has actually moved based on the target value;
Correcting the movement trajectory based on the movement trajectory and the actual movement trajectory;
A method for controlling a moving body
　１００　　行動計画部
　１２８　　移動軌跡補正部
　２００　　制御部
　２２０　　実移動軌跡取得部 DESCRIPTION OF SYMBOLS 100 Action plan part 128 Movement locus correction part 200 Control part 220 Actual movement locus acquisition part

Claims (8)

  1. 　移動体を制御するための目標値と、前記目標値から予測される前記移動体の移動軌跡を計画する行動計画部と、
    　前記目標値に基づいて前記移動体が実際に移動した実移動軌跡を取得する実移動軌跡取得部と、
    　前記移動軌跡と前記実移動軌跡に基づいて、前記移動軌跡を補正する移動軌跡補正部と、
    　を備える、移動体の制御装置。     A target value for controlling the moving object, and an action planning unit that plans a movement trajectory of the moving object predicted from the target value;
    An actual movement trajectory acquisition unit that acquires an actual movement trajectory in which the moving body has actually moved based on the target value;
    A movement locus correction unit for correcting the movement locus based on the movement locus and the actual movement locus;
    A control device for a moving body.
  2. 　前記行動計画部は、複数の前記目標値と複数の前記目標値のそれぞれから予測される前記移動軌跡を計画し、
    　前記移動軌跡補正部は、特定の前記目標値に対応する前記実移動軌跡に基づいて、前記特定の前記目標値から予測される前記移動軌跡を補正する、請求項１に記載の移動体の制御装置。     The action planning unit plans the movement trajectory predicted from each of the plurality of target values and the plurality of target values,
    2. The control of the moving body according to claim 1, wherein the movement trajectory correction unit corrects the movement trajectory predicted from the specific target value based on the actual movement trajectory corresponding to the specific target value. apparatus.
  3. 　前記移動軌跡補正部は、前記特定の前記目標値から予測される前記移動軌跡を前記実移動軌跡に補正する、請求項２に記載の移動体の制御装置。 3. The moving body control device according to claim 2, wherein the movement trajectory correction unit corrects the movement trajectory predicted from the specific target value to the actual movement trajectory.
  4. 　前記行動計画部は、複数の前記目標値と複数の前記目標値のそれぞれから予測される前記移動軌跡を計画し、
    　前記移動軌跡補正部は、特定の前記目標値に対応する前記実移動軌跡に基づいて、他の前記目標値から予測される前記移動軌跡を補正する、請求項１に記載の移動体の制御装置。     The action planning unit plans the movement trajectory predicted from each of the plurality of target values and the plurality of target values,
    2. The moving body control device according to claim 1, wherein the movement trajectory correction unit corrects the movement trajectory predicted from another target value based on the actual movement trajectory corresponding to the specific target value. .
  5. 　前記移動軌跡補正部は、現制御周期より以前の制御周期における前記移動軌跡と前記実移動軌跡に基づいて、現制御周期における前記移動軌跡を補正する、請求項１に記載の移動体の制御装置。 2. The moving body control device according to claim 1, wherein the movement trajectory correction unit corrects the movement trajectory in a current control cycle based on the movement trajectory and the actual movement trajectory in a control cycle prior to the current control cycle. .
  6. 　前記目標値に基づいて前記移動体を制御する制御部を備える、請求項１に記載の移動体の制御装置。 The control apparatus of the moving body of Claim 1 provided with the control part which controls the said moving body based on the said target value.
  7. 　前記行動計画部は、大域的な行動計画を行うグローバルパスプランニング部と、局所的な行動計画を行うローカルパスプランニング部とを含み、
    　前記目標値と、前記目標値から予測される前記移動体の移動軌跡は、前記グローバルパスプランニング部による前記大域的な行動計画に基づいて、前記ローカルパスプランニング部が計画する、請求項１に記載の移動体の制御装置。     The action planning unit includes a global path planning unit that performs a global action plan, and a local path planning unit that performs a local action plan,
    The local path planning unit plans the target value and the movement trajectory of the mobile body predicted from the target value based on the global action plan by the global path planning unit. Mobile body control device.
  8. 　移動体を制御するための目標値と、前記目標値から予測される前記移動体の移動軌跡を計画することと、
    　前記目標値に基づいて前記移動体が実際に移動した実移動軌跡を取得することと、
    　前記移動軌跡と前記実移動軌跡に基づいて、前記移動軌跡を補正することと、
    　を備える、移動体の制御方法。     Planning a target value for controlling the moving body, and a movement trajectory of the moving body predicted from the target value;
    Acquiring an actual movement trajectory in which the moving body has actually moved based on the target value;
    Correcting the movement trajectory based on the movement trajectory and the actual movement trajectory;
    A method for controlling a moving body.
PCT/JP2019/002716 2018-02-28 2019-01-28 Mobile body control device and mobile body control method WO2019167511A1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
JP2018-034295 2018-02-28
JP2018034295A JP2021071734A (en) 2018-02-28 2018-02-28 Control device for moving body and control method for moving body

Publications (1)

Publication Number Publication Date
WO2019167511A1 true WO2019167511A1 (en) 2019-09-06

Family

ID=67805607

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2019/002716 WO2019167511A1 (en) 2018-02-28 2019-01-28 Mobile body control device and mobile body control method

Country Status (2)

Country Link
JP (1) JP2021071734A (en)
WO (1) WO2019167511A1 (en)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110687923A (en) * 2019-11-08 2020-01-14 深圳市道通智能航空技术有限公司 Unmanned aerial vehicle long-distance tracking flight method, device, equipment and storage medium
WO2020203342A1 (en) * 2019-04-02 2020-10-08 ソニー株式会社 Control device, control method, and program
WO2020203341A1 (en) * 2019-04-02 2020-10-08 ソニー株式会社 Control device, control method, and program
CN112327848A (en) * 2020-11-05 2021-02-05 北京京东乾石科技有限公司 Robot obstacle avoidance method and device, storage medium and electronic equipment

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH02110606A (en) * 1988-10-19 1990-04-23 Robotetsuku Kenkyusho:Kk Remote controlling system for moving body
JP2008201177A (en) * 2007-02-16 2008-09-04 Toyota Motor Corp Parking assist device
JP2009012672A (en) * 2007-07-06 2009-01-22 Toyota Motor Corp Traveling controller for vehicle
JP2010055244A (en) * 2008-08-27 2010-03-11 Pioneer Electronic Corp Safety support device, safety support system, and safety support method

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH02110606A (en) * 1988-10-19 1990-04-23 Robotetsuku Kenkyusho:Kk Remote controlling system for moving body
JP2008201177A (en) * 2007-02-16 2008-09-04 Toyota Motor Corp Parking assist device
JP2009012672A (en) * 2007-07-06 2009-01-22 Toyota Motor Corp Traveling controller for vehicle
JP2010055244A (en) * 2008-08-27 2010-03-11 Pioneer Electronic Corp Safety support device, safety support system, and safety support method

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
SUGANUMA NAOKI ET AL.: "2.2 OVERVIEW OF AUTOMATIC DRIVING SYSTEM", THE 31ST ANNUAL CONFERENCE OF THE ROBOTICS SOCIETY OF JAPAN, September 2013 (2013-09-01) *

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2020203342A1 (en) * 2019-04-02 2020-10-08 ソニー株式会社 Control device, control method, and program
WO2020203341A1 (en) * 2019-04-02 2020-10-08 ソニー株式会社 Control device, control method, and program
CN110687923A (en) * 2019-11-08 2020-01-14 深圳市道通智能航空技术有限公司 Unmanned aerial vehicle long-distance tracking flight method, device, equipment and storage medium
CN110687923B (en) * 2019-11-08 2022-06-17 深圳市道通智能航空技术股份有限公司 Unmanned aerial vehicle long-distance tracking flight method, device, equipment and storage medium
CN112327848A (en) * 2020-11-05 2021-02-05 北京京东乾石科技有限公司 Robot obstacle avoidance method and device, storage medium and electronic equipment

Also Published As

Publication number Publication date
JP2021071734A (en) 2021-05-06

Similar Documents

Publication Publication Date Title
WO2019167511A1 (en) Mobile body control device and mobile body control method
CN108473132B (en) Parking assistance system, parking assistance method, and program
Li et al. Development of a new integrated local trajectory planning and tracking control framework for autonomous ground vehicles
KR101145112B1 (en) Steering control device of autonomous vehicle, autonomous vehicle having the same and steering control method of autonomous vehicle
WO2018079069A1 (en) Vehicle control device
US20200159216A1 (en) Motion Planning Methods And Systems For Autonomous Vehicle
CN110361013B (en) Path planning system and method for vehicle model
WO2018084324A1 (en) Method and system for controlling vehicle
WO2018061612A1 (en) Vehicle control device
Ritschel et al. Nonlinear model predictive path-following control for highly automated driving
CN113165648A (en) Control system and control method for planning possible trajectories of a motor vehicle on the basis of sampling
JP2017077849A (en) Vehicle traveling control device
RU2749529C1 (en) Vehicle control system
US10474149B2 (en) Autonomous behavior control using policy triggering and execution
JP6419671B2 (en) Vehicle steering apparatus and vehicle steering method
Artuñedo et al. Jerk-limited time-optimal speed planning for arbitrary paths
Kim et al. Vehicle path prediction using yaw acceleration for adaptive cruise control
CN113286732A (en) Control system and control method for hybrid method for determining possible trajectories of a motor vehicle
JP2015036840A (en) Autonomous travel vehicle, control system for said vehicle, control method for said vehicle
JP2009053926A (en) Route-planning device and route-planning method
JP4467533B2 (en) Folding line following mobile robot and control method of broken line following mobile robot
WO2022217209A1 (en) Lane changing based only on local information
CN110626340A (en) Intelligent automobile path tracking control system and method based on wolf algorithm
WO2019003302A1 (en) Vehicle control apparatus
Ziegler et al. Anytime tree-based trajectory planning for urban driving

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19761397

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19761397

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: JP