KR20220097756A - Driving method or road designing method for preventing dizziness for passengers - Google Patents
Driving method or road designing method for preventing dizziness for passengers Download PDFInfo
- Publication number
- KR20220097756A KR20220097756A KR1020200189045A KR20200189045A KR20220097756A KR 20220097756 A KR20220097756 A KR 20220097756A KR 1020200189045 A KR1020200189045 A KR 1020200189045A KR 20200189045 A KR20200189045 A KR 20200189045A KR 20220097756 A KR20220097756 A KR 20220097756A
- Authority
- KR
- South Korea
- Prior art keywords
- vehicle
- driving
- motion sickness
- road
- trajectory
- Prior art date
Links
- 238000000034 method Methods 0.000 title claims abstract description 12
- 208000002173 dizziness Diseases 0.000 title description 3
- 201000003152 motion sickness Diseases 0.000 claims abstract description 28
- 238000005457 optimization Methods 0.000 claims abstract description 6
- 230000001133 acceleration Effects 0.000 claims description 10
- 230000002265 prevention Effects 0.000 claims description 2
- 238000013528 artificial neural network Methods 0.000 claims 1
- 238000004141 dimensional analysis Methods 0.000 claims 1
- 230000002787 reinforcement Effects 0.000 claims 1
- 230000006870 function Effects 0.000 abstract description 16
- 230000001720 vestibular Effects 0.000 description 9
- 238000005516 engineering process Methods 0.000 description 7
- 210000000056 organ Anatomy 0.000 description 6
- 230000000007 visual effect Effects 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 210000000697 sensory organ Anatomy 0.000 description 3
- 210000004556 brain Anatomy 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 210000003128 head Anatomy 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 210000002480 semicircular canal Anatomy 0.000 description 2
- 241000406668 Loxodonta cyclotis Species 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 238000012888 cubic function Methods 0.000 description 1
- 239000003814 drug Substances 0.000 description 1
- 229940079593 drug Drugs 0.000 description 1
- 210000001508 eye Anatomy 0.000 description 1
- 150000002500 ions Chemical class 0.000 description 1
- 210000004880 lymph fluid Anatomy 0.000 description 1
- 210000005036 nerve Anatomy 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 230000011514 reflex Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000004043 responsiveness Effects 0.000 description 1
- 210000005077 saccule Anatomy 0.000 description 1
- 230000001953 sensory effect Effects 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 230000000638 stimulation Effects 0.000 description 1
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W60/00—Drive control systems specially adapted for autonomous road vehicles
- B60W60/001—Planning or execution of driving tasks
- B60W60/0013—Planning or execution of driving tasks specially adapted for occupant comfort
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W30/00—Purposes of road vehicle drive control systems not related to the control of a particular sub-unit, e.g. of systems using conjoint control of vehicle sub-units
- B60W30/10—Path keeping
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W30/00—Purposes of road vehicle drive control systems not related to the control of a particular sub-unit, e.g. of systems using conjoint control of vehicle sub-units
- B60W30/14—Adaptive cruise control
- B60W30/143—Speed control
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W40/00—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models
- B60W40/08—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models related to drivers or passengers
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W40/00—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models
- B60W40/10—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models related to vehicle motion
- B60W40/105—Speed
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W60/00—Drive control systems specially adapted for autonomous road vehicles
- B60W60/001—Planning or execution of driving tasks
- B60W60/0015—Planning or execution of driving tasks specially adapted for safety
- B60W60/0016—Planning or execution of driving tasks specially adapted for safety of the vehicle or its occupants
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W40/00—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models
- B60W40/08—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models related to drivers or passengers
- B60W2040/0818—Inactivity or incapacity of driver
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W50/00—Details of control systems for road vehicle drive control not related to the control of a particular sub-unit, e.g. process diagnostic or vehicle driver interfaces
- B60W2050/0001—Details of the control system
- B60W2050/0002—Automatic control, details of type of controller or control system architecture
- B60W2050/0013—Optimal controllers
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W50/00—Details of control systems for road vehicle drive control not related to the control of a particular sub-unit, e.g. process diagnostic or vehicle driver interfaces
- B60W2050/0001—Details of the control system
- B60W2050/0019—Control system elements or transfer functions
- B60W2050/0021—Differentiating means
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W2300/00—Indexing codes relating to the type of vehicle
- B60W2300/18—Four-wheel drive vehicles
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W2520/00—Input parameters relating to overall vehicle dynamics
- B60W2520/10—Longitudinal speed
- B60W2520/105—Longitudinal acceleration
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60Y—INDEXING SCHEME RELATING TO ASPECTS CROSS-CUTTING VEHICLE TECHNOLOGY
- B60Y2300/00—Purposes or special features of road vehicle drive control systems
- B60Y2300/10—Path keeping
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60Y—INDEXING SCHEME RELATING TO ASPECTS CROSS-CUTTING VEHICLE TECHNOLOGY
- B60Y2300/00—Purposes or special features of road vehicle drive control systems
- B60Y2300/14—Cruise control
- B60Y2300/143—Speed control
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60Y—INDEXING SCHEME RELATING TO ASPECTS CROSS-CUTTING VEHICLE TECHNOLOGY
- B60Y2400/00—Special features of vehicle units
- B60Y2400/82—Four wheel drive systems
Landscapes
- Engineering & Computer Science (AREA)
- Automation & Control Theory (AREA)
- Transportation (AREA)
- Mechanical Engineering (AREA)
- Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Human Computer Interaction (AREA)
- Theoretical Computer Science (AREA)
- Biomedical Technology (AREA)
- General Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Traffic Control Systems (AREA)
Abstract
Description
인공지능기술의 발달은 자동차의 무인화기술에 새로운 분야로 자리잡고 있다. 실제로 웹서비스로 진행되는 자율주행 차량학습 사이트에서도 차량의 중간선부터 이탈거리, 차량의 방향각도, 조향각도 등을 이용해 다양한 요인을 점수로서 학습시킬 수 있다. The development of artificial intelligence technology is positioned as a new field in unmanned automobile technology. In fact, in the self-driving vehicle learning site that is conducted as a web service, various factors can be learned as scores by using the distance from the midline of the vehicle, the direction angle of the vehicle, and the steering angle.
운전자는 타 탑승자에 비해 멀미를 덜 느낀다. 차멀미의 주 원인은 시각정보와 전정기관의 정보차이가 느껴 지기 때문이다. 운전자는 차량의 물리적특성을 조종하고 있기에 시각정보와 전정기관의 정보를 미리 예측할 수 있다. 이로서 정보의 차는 줄어들 수 있고 멀미는 덜 느끼는 것이다. 동승자가 운전자의 운전습관을 좀더 학습하면 동승자 역시 멀미가 줄어들 수 있으며 이를 위해 조명장치를 통해 동승자를 학습시키는 장치가 연구된 바도 있다. Drivers feel less motion sickness than other passengers. The main cause of motion sickness is that the difference between visual information and vestibular organ information is felt. Since the driver controls the physical characteristics of the vehicle, visual information and information of the vestibular system can be predicted in advance. In this way, the information gap can be reduced and motion sickness is less felt. If the passenger learns more about the driver's driving habits, the passenger can also reduce motion sickness.
만약 자율주행기술이 탑재된다면 학습 해야 할 운전습관이 한가지 일 것이므로 자율주행자체가 멀미방지에 도움이 될 수 있다. 또한, 멀미방지를 최적화 시킨 운전습관을 학습시킬 수 있다. 보고된 바에 의하면 시각장애인도 멀미를 느낀다. 이는 시각정보가 없이 전정기관의 작용만으로도 멀미를 충분히 느낄 수 있다는 뜻이다. 그래서 이를 방지하기 위한 장치들이 연구개발되곤 한다. 한 예로 Galvanic Vestibular Stimulation 장치는 전정기관에 음이온을 방출하여 전정기관에 신호를 인공적으로 넣음 으로서 가속을 체험하고 있다고 신경을 이용해 뇌에 착각을 일으키는 장치이다. 이를 머리에 차고 자동차에 타서 멀미를 줄이는 장치에 대한 연구도 나왔다. 허나 전정기관단위에서 전위를 주게 되면 뇌 뿐만 아니라 반사신경도 같이 반응하게 되어 몸이 스스로 균형을 잡으려고 기울어지거나 눈이 자동으로 회전할 수 있다. 그리고 무엇보다도 장치를 구매 및 착용해야 한다. 문제의 본질적으로 돌아와서 처음부터 운전습관만 최적화 되어 있다면 타장치를 사용해야할 필요성도 없어질 것이므로 해당기술은 큰 돌파구가 될 수 있다. If autonomous driving technology is installed, there will be one driving habit to learn, so autonomous driving itself can help prevent motion sickness. In addition, it is possible to learn driving habits optimized for motion sickness prevention. Reportedly, blind people also experience motion sickness. This means that without visual information, motion sickness can be sufficiently felt by the action of the vestibular system. Therefore, devices to prevent this are often researched and developed. For example, the Galvanic Vestibular Stimulation device emits negative ions to the vestibular organ and artificially inserts a signal into the vestibular organ to create an illusion in the brain that it is experiencing acceleration. There has also been a study on a device to reduce motion sickness by wearing it on the head and riding in a car. However, when an electric potential is given from the vestibular system unit, not only the brain but also the reflexes react, so that the body can tilt to balance itself or the eyes can automatically rotate. And above all else, you have to buy and wear the device. Returning to the essence of the problem, if driving habits are optimized from the beginning, there will be no need to use other devices, so this technology can be a big breakthrough.
한 연구기관에 따르면 다른 가속도에 한시간 이상 노출될 경우에 이에 적응하기 위해 멀미를 느낄 수 있다. 즉, 비교적 순간적인 차량 탑승시에서는 선형 가속도의 크기가 중요한 것이 아니다. 선형 가속의 변화율이 중요한 것이다. According to one research institute, if you are exposed to different accelerations for more than an hour, you may feel motion sick to adapt to them. That is, the magnitude of the linear acceleration is not important when riding in a relatively instantaneous vehicle. The rate of change of the linear acceleration is important.
우리가 멀미를 느끼는 주 이유는 시각정보와 균형감각정보의 차로 인해 생긴다. 몸이 회전을 멈췄는데 림프액은 계속회전해서 시각정보와의 오차로 인해 멀미가 발생하는 것이다. 코끼리코를 도는 놀이를 할 때 끝나고 멈추었을 때 특히 어지러움을 느껴볼 수 있다. The main reason we feel motion sickness is caused by the difference between visual information and balance sensory information. The body stops rotating, but the lymph fluid continues to rotate, resulting in motion sickness due to an error with the visual information. When playing around the elephant's nose, you may feel particularly dizzy when it is finished and stopped.
자율주행의 주행경로 및 속도를 적당히 주는 차량의 궤적 또는 차도의 궤도를 설계를 통해 탑승자의 멀미 및 차량안의 물체에 대한 흔들림을 줄이고자 한다.It is intended to reduce motion sickness of occupants and shaking of objects in the vehicle by designing the trajectory of the vehicle or the trajectory of the road that gives an appropriate driving path and speed for autonomous driving.
차가 직선경로에서 곡선경로를 진입할 때 순간적으로 곡률이 무한대에서 일정 상수로 진입할 수는 없다. 물론 곡선 진입로(위 도면에서의 동그라미)에서 모형차를 멈추고 서보를 돌린 뒤 다시 출발한다면 가능할 것이다. 수 있다. 즉, 정지 없이 곡선경로를 진입하는 완벽한 궤적은 없다. When a car enters a curved path from a straight path, the curvature cannot instantaneously enter from infinity to a constant constant. Of course, it will be possible if you stop the model car on a curved ramp (circle in the drawing above), turn the servo, and start again. can That is, there is no perfect trajectory that enters a curved path without stopping.
모형차의 회전중심이 뒷바퀴 회전축 위에 존재한다고 가정하였을 때에 차량이 그리는 궤도는 그림과 같다. 도 4에서 safety factor를 차량과 차선과의 거리라고 했을 때 양쪽 차선에 대한 거리가 같을 때 가장 안전하게 주행한다고 할 수 있으며 이는 불연속적인 것을 알 수 있다. 따라서 차량주행에 있어서 정답은 있을 수 없으며 최적화라는 작업이 필요하다. 물론 safety factor 혹은 곡률차(현재곡률 과 도로 곡률의 차) 도 최적화하는데 보상함수 등으로 적용할 수 있다 Assuming that the center of rotation of the model car is on the rear wheel axis, the trajectory drawn by the vehicle is as shown in the figure. In FIG. 4 , when the safety factor is the distance between the vehicle and the lane, it can be said that the vehicle is driven most safely when the distance to both lanes is the same, and it can be seen that this is discontinuous. Therefore, there is no right answer in vehicle driving, and the task of optimization is necessary. Of course, the safety factor or curvature difference (the difference between the current curvature and the curvature of the road) can also be optimized and applied as a compensation function.
차량의 동역학적 특성을 분석하여 멀미를 일으키는 주 요인들을 수치화 하여 cost function, reward function 등으로 표현하여 동적프로그래밍, 대안 컴퓨팅 등의 학습 및 최적화된 시킨 차량의 궤적 또는 차도의 궤도를 도출해 낸다.By analyzing the dynamic characteristics of the vehicle, the main factors that cause motion sickness are digitized and expressed as a cost function, a reward function, etc. to derive the trajectory or the trajectory of the vehicle that has been optimized for learning and optimization such as dynamic programming and alternative computing.
4차산업혁명시대의 성과물로 자율주행이 급부상 하고 있다. 이로서 사람이 직접 운전하는 상황에서의 단점을 보강할 수 있다. 차멀미는 운전자의 습관에 따라서 멀미가 발생하는 정도가 다르다. 이러한 부분을 기술의 발달을 이용하여 보강할 수 있다. 현재는 멀미자체를 배제하려고 다양한 시도가 있었다. 이제는 자율주행기술을 이용하여 주행자체가 멀미를 나지 않도록 할 수 있다. Autonomous driving is rapidly emerging as a result of the 4th industrial revolution era. In this way, it is possible to reinforce the disadvantages of a situation in which a person is directly driving. The degree of motion sickness varies depending on the driver's habits. These parts can be reinforced using the development of technology. Currently, various attempts have been made to exclude motion sickness itself. Now, by using autonomous driving technology, driving itself can prevent motion sickness.
더이상 차멀미를 위해서 의학적 약품섭취 혹은 의학장치 부착 등의 노력이 필요 없어질 것이다. 또한 차멀미를 싫어해서 차를 타지 않는 사람들에게도 긍정적으로 차에게 한발 다가가는 계기가 될 수 있다.It will no longer be necessary to take medical drugs or attach medical devices for car sickness. Also, it can be an opportunity to take a positive approach to the car for those who do not ride a car because they hate car sickness.
더불어 차량속에 있는 짐들에 대해서도 보다 안정적인 환경을 유지할 수 있을 것이다. 짐이 벽에 계속다시부딪히고 넘어지고 뒤집히면 손상등의 불편이 있을 수 있다.In addition, it will be possible to maintain a more stable environment for luggage in the vehicle. If the luggage keeps hitting the wall again, falling over, and overturning, there may be inconveniences such as damage.
도 1은 인간의 신체중 귀에 있는 평형기관을 나타내는 도면이다. 전정기관(vestibular organ)은 우리의 몸이 균형을 측정하기위한 몸의 한 기관이다. 둥근주머니(Saccule)는 수직 방향의 가속을, 타원주머니(Utricle)는 수평 방향의 가속을, 세개의 반고리관은 : 세가지 방향의 회전 움직임을 감지한다.
도 2 는 Galvanic Vestibular Stimulator 장치에 대한 그림으로서, 머리에 밴드를 착용하고 밴드를 통해 신경에 착시 신호를 보내는 장치이다.
도 3 은 차량이 가지는 회전중심 및 곡률에 대해서 설명한 것이다. 도로의 회전곡률은 불연속적임을 표현한 도면이다.
도 4 는 도로에서 차량과 차선사이의 거리를 최소한으로 하는 차량궤적은 불연속적이라는 것을 표현한 것이다.
도 5 는
자유도는 (Θ,Δr, ρ, Θs) 네가지 이다.
Θ : 도로와 모형차의 사잇각
Δr : 모형차의 중심부터 최적 궤도의 거리
ρ : 최적 궤도의 곡률
s : 서보모터 각도
모형차를 제어목적으로 분석하기 위해 Θ 혹은 Δr 또는 둘다 를 입력으로 받고 Θs 를 출력으로 넣을 수 있다.
도 6은 주행도로 및 탑승장치가 가지는 특성들을 도면상에 표현한 것이다.
도 7은 상기 도6의 주행도로와 탑승장치가 한곳에 모여있는 도면이며 탑승장치의 회전중심과 도로의 회전중심이 일치하지 않는 순간이다.1 is a view showing a balance organ in the ear of the human body. The vestibular organ is an organ in the body for which our body measures balance. The saccule senses vertical acceleration, the utricle senses horizontal acceleration, and the three semicircular canals: detect rotational movement in three directions.
FIG. 2 is a diagram of a Galvanic Vestibular Stimulator device, which is a device that wears a band on the head and sends an optical illusion signal to the nerve through the band.
3 illustrates a rotation center and curvature of a vehicle. It is a diagram expressing that the turning curvature of the road is discontinuous.
4 illustrates that the vehicle trajectory that minimizes the distance between the vehicle and the lane on the road is discontinuous.
5 is
There are four degrees of freedom (Θ,Δr, ρ, Θs).
Θ : The angle between the road and the model car
Δr : Distance of the optimal trajectory from the center of the model car
ρ: the curvature of the optimal trajectory
s : servo motor angle
To analyze the model difference for control purposes, we can take Θ or Δr or both as input and put Θs as output.
6 is a diagram illustrating characteristics of a driving road and a vehicle.
7 is a view in which the driving road and the vehicle of FIG. 6 are gathered in one place, and is a moment when the rotational center of the vehicle and the rotational center of the road do not coincide.
차량이 항상 중심점을 기준으로 회전운동을 한다고 가정하고 이를 극좌표로 해석해 볼 수 있다. 반지름 r, 각도 는 어느 탑승객 위치를 중심으로 잡느냐에 따라 변경할 수 있다. 보조석 혹은 후방석 등 멀미에 특히 취약한 탑승객의 자리를 기준으로 최적화시킬 수 있겠다. 아래와 같이 동역학적 수식으로 표현할 수 있다.Assuming that the vehicle always rotates with respect to the center point, it can be interpreted as polar coordinates. radius r, angle can be changed depending on which passenger position is centered. It can be optimized based on the seat of passengers who are particularly vulnerable to motion sickness, such as the passenger seat or the rear seat. It can be expressed as a dynamic equation as follows.
, 또는 등의 값이 중요하게 작용하며 이는 우리의 감각기관중 선형적인 방향의 가속에 관련된 기관들이 있기 때문이다. 이값이 크면 멀미가 더많이 날 것이다. , or The value of etc. is important because some of our sensory organs are related to acceleration in a linear direction. The higher this value, the more motion sickness will occur.
여기서 r은 핸들각도로 인해 정해지는 각도이며, 는 r 과 v 즉 핸들과 v 에 관련된 함수이다.where r is the angle determined by the handle angle, is a function related to r and v, that is, a handle and v .
전력 P를 나타내어 전개하면 아래와같다.The power P is expressed and expanded as follows.
이를 통해 전력 P와 핸들각들을 입력으로 해석할 수 있다.Through this, power P and steering wheel angles can be interpreted as inputs.
회전하는 자동차를 태양으로 보고, 자동차안의 탑승객의 회전감각기관을 행성으로서 해석하고 아래와같이 나타낼 수 있다.The rotating car can be viewed as the sun, and the rotation sensory organs of the passengers in the car can be interpreted as planets and expressed as follows.
Planer p, sun sPlaner p, sun s
관성모멘트 공식으로 시작한다.We start with the moment of inertia formula.
| 의 값이 중요하며 이는 우리의 감각기관중 회전방향의 가속에 관련된 기관들이 있기 때문이다. 마찬가지로 이값이 크면 멀미가 더많이 날 것이다.| The value of is important because some of our sensory organs are involved in the acceleration of the rotational direction. Likewise, the higher this value, the more motion sickness will occur.
이러한 변수들; ,| 등을 사용하여 최적함수 즉, cost function J or reward function R, 등에 사용해 최적화를 시킬 수 있다.these variables; ,| It can be optimized by using the optimal function, i.e., cost function J or reward function R, etc.
: A : A
: B : B
: C : C
7 control variables 7 control variables
위와같이 3개의 식과 7개의 control variable 이 있으며 이는 선형대수적으로 무한대의 네제곱 상당의 가능한 방법이 있다는 것이다. As above, there are 3 equations and 7 control variables, which means that there are possible methods equivalent to the fourth power of infinity in linear algebra.
율주행에 사용할 수 있는 모델을 제안한다. We propose a model that can be used for self-driving.
가로축이 입력으로 Δr, 세로축이 출력으로 s 라고 하면 아래 위와 같이 제어모델을 그려볼 수 있다. 여기서 원점은 ( ρ, (ρ)[ s] ) 임을 주의하자. If the horizontal axis is Δr as the input and the vertical axis is s as the output, the control model can be drawn as shown below. Note that the origin is ( ρ, (ρ)[ s] ).
형태의 기함수로 표현할 수 있다. 기함수인 이유는 주행이 좌우 대칭해야하기 때문이다. 갈색의 함수는 입력값에 따라 경우의 수로 나눠서 제어하는 함수이다. 삼차함수로 나타낸 이유는 기함수 테일러 급수의 두번 째 항까지만 사용한 것이다. It can be expressed as an odd function of the form. The reason for the odd function is that the driving must be symmetrical. The brown function is a function that is controlled by dividing the number of cases according to the input value. The reason why it is expressed as a cubic function is that only the second term of the odd Taylor series is used.
Δr 만 입력으로 받은 경우에는 카메라에서 다가오는 차도를 보아도 그에 미리 대응할 수 없다. 그래프로 그려보면 위와 같다. 이를 가능하게 하려면 다가오는 차도의 최적궤도의 정보를 미리 파악하여 계산하는 함수 g(t)가 필요하다. g(t)는 본인의 위치인 0에서부터 특정 거리 α 까지 적분을 해야 한다. If only Δr is received as an input, even if the camera sees the approaching road, it is not possible to respond in advance. As a graph, it looks like the above. To make this possible, a function g(t) that calculates the information on the optimal trajectory of the approaching roadway is required in advance. g(t) must be integrated from 0, which is your position, to a specific distance α.
만약 위와 같은 delta function 이 측정된다고 해보자. 방향을 돌리는 정도는 시간에 따라서 더 깊어지는게 좋을 것이다. 그러기 위해선 delta function 이 원점에 가까울수록 가중치가 커져야한다. 따라서 α - s에 관한 항을 넣는다. Suppose that the above delta function is measured. It would be better if the degree of turning the direction deepens with time. To do this, the closer the delta function is to the origin, the greater the weight should be. Therefore, we put the term for α - s.
최종식 : Final expression:
기존의 방식으로는 제어방식을 PID를 사용하여 제어하는 연구가 많다. 물론 여기서도 제어하고 싶은 변수를 조정 해 볼 수 있으며 위의 블록선도와 같이 forecasting 알고리즘을 탑재해 볼 수 있을 것이다.In the existing method, there are many studies that control the control method using PID. Of course, you can also try adjusting the variables you want to control here, and you can try loading the forecasting algorithm like the block diagram above.
- 주행의 최적화- Optimization of driving
만약 최적화가 잘 되지 않아서 반응도가 너무 크다면 위와 같이 단순한 직선에서도 경로를 이탈해버릴 수도 있을 것이다. 따라서 이 상수값들을 최적화 해줄 필요가 있다. If the responsiveness is too high due to poor optimization, it may deviate from the path even in a simple straight line like the one above. Therefore, it is necessary to optimize these constant values.
여기서 최적화 되어야 할 상수들을 우리가 물리적 즉, 직접적으로 시연을 해보면서 학습시키는 것은 매 에피소드마다 시간, 노력 등 소모되는 것이 많다. 이는 그래픽 시뮬레이션 으로 에피소드를 속전속결시켜서 학습을 시킬 수 있다. 후에 학습된 신경망 모델을 모형차에 입력시키면 최적화된 주행알고리즘을 사용할 수 있을 것이다. 학습에 차질이 생긴다면 k1,k4,k5를 제거한 뒤 하나의 상수씩 최적화 시킬 수 있으며 이는 훨씬 간단한 이분법으로 시도할 수 있다. PID 제어의 경우 KP, KI, KD 값들을 학습시켜 볼 수 있다. Here, learning the constants to be optimized by physically demonstrating them directly consumes a lot of time and effort in every episode. This is a graphic simulation that can speed up the episodes and teach them. After that, if the learned neural network model is input to the model car, the optimized driving algorithm will be available. If there is a problem in learning, k1, k4, k5 can be removed and then optimized one by one, which can be tried with a much simpler dichotomy. In the case of PID control, KP, KI, and KD values can be learned.
- Mapping- Mapping
좌측의 그림을 보면 카메라의 정보자체만으로는 이전의 궤적을 알 수 없다. 이전의 궤적이 점선처럼 직선 도로였는지, 혹은 실선처럼 곡선이였는지에 대한 정보가 없게 될 경우 Δr값의 불확실성 때문에 기구학적 제어에 차질이 생긴다. 이를 보완하기 위해 카메라에서 추출한 정보들을 기록해 둘 수있다. 이렇게 기록된 점들의 위치는 매 state마다 위치가 업데이트 되어야 할 것이며 이는 동역학적 해석으로 을 통해 추적할 수 있다. 따라서 Mapping m = m 으로 표현할 수 있으며 이는 테일러급수 일차미분까지만 표현한 것이다.Looking at the picture on the left, the previous trajectory cannot be known only from the information of the camera itself. If there is no information on whether the previous trajectory was a straight road like a dotted line or a curved road like a solid line, the kinematic control is disrupted due to the uncertainty of the Δr value. To compensate for this, information extracted from the camera can be recorded. The positions of the points recorded in this way should be updated in every state, which is a dynamic analysis. can be tracked through So Mapping m = m It can be expressed as , which is expressed only up to the first derivative of the Taylor series.
물론 주행에 필요한 방법으로는 위의 모델 말고도 카메라데이터에 대해 CNN으로 출력을 결정하는 모델 등을 사용할 수도 있다.Of course, as a method necessary for driving, in addition to the above model, a model that determines the output by CNN for camera data can be used.
차량제어부가 차량에 부착되어 판매 및 유통될 수 있다.The vehicle control unit may be attached to the vehicle and sold and distributed.
차량제어부를 구매하여 차량에 부착할 수 있다.You can purchase the vehicle control unit and attach it to the vehicle.
차량에 수신부 송신부를 장착하고 차량제어기술을 네트워크상으로 구현하여 차량에게 운전 신호를 보내줄 수 있다.By installing a receiver and transmitter in the vehicle and implementing the vehicle control technology over a network, it is possible to send a driving signal to the vehicle.
차도를 구성하는데에 있어서 사전에 가능한 도로를 컴퓨터로 최적화시켜서 사용할 수 있다.In composing a roadway, it is possible to optimize and use possible roads with a computer in advance.
100 : 탑승자 혹은 기준자로서 멀미계산의 기준이 된다.
101: 탑승장치로서 4륜구동자동차에만 국한되지 않는다.
200 : 탑승장치의 회전중심이며 큰화살표를 따라간다. 상기 부호 101과의 거리는 r 으로 표현될 수 있다. 탑승장치의 회전은 으로 표현될 수 있으며 s 는 sun이다.
201 : 탑승장치가 움직이는 방향으로 단위벡터로 표현될 수 있다.
203 : 경로가 가지는 최적궤적의 표현으로 탑승장치와 도로사이의 거리를 최소화 할 수 있는 탑승장치의 궤적으로 원궤도기반 커브길 진입로에서 불연속적인게 특징이다.
204 : 탑승자 혹은 기준자의 반고리관 속 회전체 (otholith 등) 회전으로 로 표현된다. p 는 planar으로서 상기 sun 과 구별된다.100: As a passenger or standard person, it becomes the standard for motion sickness calculation.
101: A vehicle, but not limited to a four-wheel drive vehicle.
200: It is the center of rotation of the vehicle and follows the large arrow. The distance to the
201: in the direction in which the vehicle moves It can be expressed as a unit vector.
203: This is the expression of the optimal trajectory of the path, and it is the trajectory of the vehicle that can minimize the distance between the vehicle and the road.
204: Rotation of a rotating body (otholith, etc.) in the semicircular canal of the occupant or reference person is expressed as p is a planar, distinct from the sun.
Claims (7)
Optimize driving or lane design by using information on the linear acceleration change rate or angular velocity second-order derivative of the coordinate system corresponding to a specific part of the vehicle. Here, optimization means solving Bellman equations, dynamic programming, neural network learning, optimal control LQR, reinforcement learning, etc.
In the learning of claim 1, time, path deviation, etc. can be optimized and learned as well.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
KR1020200189045A KR20220097756A (en) | 2020-12-31 | 2020-12-31 | Driving method or road designing method for preventing dizziness for passengers |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
KR1020200189045A KR20220097756A (en) | 2020-12-31 | 2020-12-31 | Driving method or road designing method for preventing dizziness for passengers |
Publications (1)
Publication Number | Publication Date |
---|---|
KR20220097756A true KR20220097756A (en) | 2022-07-08 |
Family
ID=82407315
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
KR1020200189045A KR20220097756A (en) | 2020-12-31 | 2020-12-31 | Driving method or road designing method for preventing dizziness for passengers |
Country Status (1)
Country | Link |
---|---|
KR (1) | KR20220097756A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115991186A (en) * | 2023-03-06 | 2023-04-21 | 郑州轻工业大学 | Longitudinal and transverse control method for anti-carsickness automatic driving vehicle |
CN117492453A (en) * | 2024-01-03 | 2024-02-02 | 上海友道智途科技有限公司 | Unmodeled compensation control method for batch operation vehicles |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR20180053927A (en) | 2016-11-14 | 2018-05-24 | 삼성중공업 주식회사 | Blasting apparatus |
-
2020
- 2020-12-31 KR KR1020200189045A patent/KR20220097756A/en active Search and Examination
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR20180053927A (en) | 2016-11-14 | 2018-05-24 | 삼성중공업 주식회사 | Blasting apparatus |
Non-Patent Citations (3)
Title |
---|
G-Shock Trainer Lesson Plan: |
What Causes Dizziness? SPACEPORT AMERICA |
우주인의 ‘우주멀미’는 왜 생길까 The ScineceTimes |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115991186A (en) * | 2023-03-06 | 2023-04-21 | 郑州轻工业大学 | Longitudinal and transverse control method for anti-carsickness automatic driving vehicle |
CN115991186B (en) * | 2023-03-06 | 2023-08-11 | 郑州轻工业大学 | Longitudinal and transverse control method for anti-carsickness automatic driving vehicle |
CN117492453A (en) * | 2024-01-03 | 2024-02-02 | 上海友道智途科技有限公司 | Unmodeled compensation control method for batch operation vehicles |
CN117492453B (en) * | 2024-01-03 | 2024-04-12 | 上海友道智途科技有限公司 | Unmodeled compensation control method for batch operation vehicles |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US10134279B1 (en) | Systems and methods for visualizing potential risks | |
EP3303025B1 (en) | Methods and systems for controlling vehicle body motion and occupant experience | |
US9904287B1 (en) | Systems and methods for mitigating vigilance decrement while maintaining readiness using augmented reality in a vehicle | |
Amer et al. | Adaptive modified Stanley controller with fuzzy supervisory system for trajectory tracking of an autonomous armoured vehicle | |
KR20220097756A (en) | Driving method or road designing method for preventing dizziness for passengers | |
US20190101930A1 (en) | Impedance-based motion control for autonomous vehicles | |
US10551835B2 (en) | Systems and methods for operator skill mitigation | |
CN106573377A (en) | Humanoid robot with collision avoidance and trajectory recovery capabilities | |
Moret | Dynamic modeling and control of a car-like robot | |
CN108897216A (en) | A kind of intelligent automobile model prediction decision-making technique based on the virtual field of force in region | |
Li et al. | Mitigating motion sickness in automated vehicles with frequency-shaping approach to motion planning | |
Yan et al. | Driver’s individual risk perception-based trajectory planning: A human-like method | |
Chen et al. | A new lane keeping method based on human-simulated intelligent control | |
WO2020252090A1 (en) | Method and apparatus for trajectory shape generation for autonomous vehicles | |
US20220242449A1 (en) | Vehicle Control System | |
Jamson | Motion cueing in driving simulators for research applications | |
Wang et al. | Potential field based path planning with predictive tracking control for autonomous vehicles | |
Young | Robotics in the real world: the perceptual control theory approach | |
Hegedüs et al. | Hybrid trajectory planning for autonomous vehicles using neural networks | |
Rañó et al. | Naturalistic lane-keeping based on human driver data | |
Czygier et al. | Autonomous Searching Robot with Object Recognition Based on Neural Networks | |
Zhang et al. | Path planning algorithm of the intelligent vehicle considering the passenger feelings | |
Vasquez et al. | An open framework for human-like autonomous driving using Inverse Reinforcement Learning | |
Liu et al. | Car-like mobile robot reverse parking: Using fuzzy logic control approach | |
Prokop | Model of human vehicle driving-a predictive nonlinear optimization approach |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
A201 | Request for examination |