Autonomous vehicles are currently one of the most interesting smart car technologies, and have been developing by many automobile companies in the world.
Every autonomous vehicle has to be equipped with not only automatic guidance controller but also automatic velocity controller. It is necessary ...
Autonomous vehicles are currently one of the most interesting smart car technologies, and have been developing by many automobile companies in the world.
Every autonomous vehicle has to be equipped with not only automatic guidance controller but also automatic velocity controller. It is necessary for one to get dynamic models between acceleration/brake pedal inputs and driving/braking torques at vehicle wheels. in order to design an automatic velocity control system.
In this thesis we deal with a practical modeling of the acceleration and brake systems of an autonomous vehicle. The acceleration system here is represented by the transfer function model from acceleration pedal command input to the front wheel force. The model, , is identified using the moments of a single rectangular pulse response data. The pulse response data can be obtained through experimental tests. In this thesis, a real autonomous vehicle that has been rearranged by a SUV (Tucson ix, 2010, Hyundai Motor Co.) was used. Driving and braking torques of the front wheels was measured by the wheel force transducer (SWIFT 30, MTS). Several acceleration pedal commands of amplitude 40%, 50%, and 60% were considered. By means of the method of moment matching developed by Kim, we identified the acceleration system models of the second order plus zero with delay free (SOZDF) for . To validate the model,, we first combined the acceleration model with the 16 DOF full vehicle dynamic model, FVDM16 [13], and its responses are compared with the measured speeds. As a result, it is shown that the all resulting identified models have been well fitted to the measured data. The brake system here is represented by transfer function model, , from the brake pedal command input to the front wheel force. The same method is used to identify the models of brake system. Experiments were performed at two different speeds of 40km/h, 60km/h. We consider three brake pedal commands of amplitude 30%, 50%, and 70%. Applying the method of moment matching by Kim yields the brake system models; and . Similar to the acceleration model, the braking models have been validated by using the FVDM16 and a set of measured data. It is shown that the all resulting identified models are well fitted to the measured data.
Concludingly, the input-output modeling approach for the acceleration and brake systems of autonomous vehicle can be used for practical applications.
Autonomous vehicles are currently one of the most interesting smart car technologies, and have been developing by many automobile companies in the world.
Every autonomous vehicle has to be equipped with not only automatic guidance controller but also automatic velocity controller. It is necessary for one to get dynamic models between acceleration/brake pedal inputs and driving/braking torques at vehicle wheels. in order to design an automatic velocity control system.
In this thesis we deal with a practical modeling of the acceleration and brake systems of an autonomous vehicle. The acceleration system here is represented by the transfer function model from acceleration pedal command input to the front wheel force. The model, , is identified using the moments of a single rectangular pulse response data. The pulse response data can be obtained through experimental tests. In this thesis, a real autonomous vehicle that has been rearranged by a SUV (Tucson ix, 2010, Hyundai Motor Co.) was used. Driving and braking torques of the front wheels was measured by the wheel force transducer (SWIFT 30, MTS). Several acceleration pedal commands of amplitude 40%, 50%, and 60% were considered. By means of the method of moment matching developed by Kim, we identified the acceleration system models of the second order plus zero with delay free (SOZDF) for . To validate the model,, we first combined the acceleration model with the 16 DOF full vehicle dynamic model, FVDM16 [13], and its responses are compared with the measured speeds. As a result, it is shown that the all resulting identified models have been well fitted to the measured data. The brake system here is represented by transfer function model, , from the brake pedal command input to the front wheel force. The same method is used to identify the models of brake system. Experiments were performed at two different speeds of 40km/h, 60km/h. We consider three brake pedal commands of amplitude 30%, 50%, and 70%. Applying the method of moment matching by Kim yields the brake system models; and . Similar to the acceleration model, the braking models have been validated by using the FVDM16 and a set of measured data. It is shown that the all resulting identified models are well fitted to the measured data.
Concludingly, the input-output modeling approach for the acceleration and brake systems of autonomous vehicle can be used for practical applications.
주제어
#자율주행차
#차량 속도 제어
#브레이크 시스템
#악셀 시스템
#시스템 모델링
#모멘트매칭
#Autonomous vehicle
#automatic vehicle speed control
#brake system
#acceleration system
#identification
#method of moment matching
#wheel force transducer(WFT)
※ AI-Helper는 부적절한 답변을 할 수 있습니다.