Key Technology Researchof Heavy Laden Vehicle Rollover Warningbased On Dynamic Prediction

Heavy vehicles with carrying capacity, high and bulky centroid such characteristicsthat make heavy vehicles on the road at high speed roll stability is not good, if heavyvehicles turning or changing lanes for emergency, vehicle rollover accident prone. Inthis paper, read a lot of domestic and foreign heavy vehicle rollover warning aboutrelated research, combined with the Youth Science Fund Project "Hierarchical HiddenMarkov Model Based HHMM heavy laden vehicle rollover warning new algorithmresearch," presents a dynamic prediction of heavy laden vehicle rollover warning keytechnology. Through the vehicle rollover study for the establishment of heavy vehiclesmoving vehicle rollover state model, through real-time monitoring of vehicle motionstate for the driver to provide rollover warning, protect the vehicle and the driver’snormal driving safety.In this paper, hidden Markov model based on the theory, the establishment of adouble hidden Markov model, the underlying hidden Markov (HMM) modelcorresponding to the vehicle motion gesture, its observation sequence is obtained bymeans of various types of vehicle sensor data; and senior hidden Markov (HMM)model corresponds to the state of motion of the vehicle and its underlying model for theobserved sequence of motion gesture recognition results. Double hidden Markov modelis able to recognize the current state of motion of the vehicle; subsequently establishedhierarchical hidden Markov model, in the hierarchical hidden Markov model, theprediction algorithm consists of three levels of HMM model algorithm combination: ie,the probability of switching driving state layer, hidden layer state (current estimatesstate vehicle condition) and the state observer layer. The hierarchical model ofinnovative hidden Markov model, hidden layers each x is defined as an HMM model,double the previous corresponding hidden Markov model is used to estimate the currenttime vehicle status; the traveling state probability switching layer based on statisticaltheory and previous time vehicle status calculating future time vehicle status switchingprobability to accurately predict future time vehicle status, where innovative citedautoregressive hidden Markov prediction method for vehicle rollover early warning ofdangerous working conditions.Through the double hidden Markov model and autoregressive hidden Markovmodel was constructed vehicle rollover joint warning algorithm, subsequently adopted Trucksim, Matlab and Simulink co-simulation of the vehicle warning algorithmsimulation to verify the feasibility of the algorithm, timeliness and accuracy of the finaltext by using microcontroller development board for authentication, combined with thevehicle rollover warning system in the vehicle controller design and experimentalverification of the ring system, further vehicle rollover warning system feasibilityauthentication.