针对汽车底盘的麋鹿测试通过性数字化判定方法研究OA

The moose test plays an instrumental role in evaluating the intelligence level of a vehicle's chassis.However,the current pass/fail judgment method relies on visual inspection,resulting in low accuracy and inability to record raw data,etc.To address this issues,this paper proposes a digital pass/fail judgment method for the moose test based on a high-precision positioning system.First,a test coordinate system was developed.A vehicle's outer contour boundary was simplified based on its pose time-series data within the test coordinate system.Considering the antenna installation position,the vehicle's motion coverage area was drawn.Then,continuous virtual road boundaries were built according to the standard requirements and high-risk collision points were identified.A quantitative metric for"vehicle-road"collision was designed.Finally,20 sets of moose tests were employed to compare this method with the visual inspection method.The reasons behind the inconsistencies in the judgment results were analyzed.It is shown that the consistency of pass/fail judgment for the proposed method is 95％,which is superior to that of the visual inspection method.