To accurately assess the safety of freeway work zones, this paper investigates the safety of vehicle lane change maneuvers with improved cellular automata model. Taking the traffic conflict and standard deviation of o...To accurately assess the safety of freeway work zones, this paper investigates the safety of vehicle lane change maneuvers with improved cellular automata model. Taking the traffic conflict and standard deviation of operating speed as the evaluation indexes, the study evaluates the freeway work zone safety. With improved deceleration probability in car-following rules and the addition of lanechanging rules under critical state, the lane-changing behavior under critical state is defined as a conflict count. Through 72 schemes of simulation runs, the possible states of the traffic flow are carefully studied. The results show that under the condition of constant saturation traffic conflict count and vehicle speed standard deviation reach their maximums when the mixed rate of heave vehicles is 40%. Meanwhile, in the case of constant heavy vehicles mix, traffic conflict count and vehicle speed standard deviation reach maximum values when saturation rate is 0.75. Integrating all simulation results, it is known the traffic safety in freeway work zones is classified into four levels： safe, relatively safe, relatively dangerous, and dangerous.展开更多
Radio modulation classification has always been an important technology in the field of communications.The difficulty of incremental learning in radio modulation classification is that learning new tasks will lead to ...Radio modulation classification has always been an important technology in the field of communications.The difficulty of incremental learning in radio modulation classification is that learning new tasks will lead to catastrophic forgetting of old tasks.In this paper,we propose a sample memory and recall framework for incremental learning of radio modulation classification.For data with different signal-to-noise ratios,we use a partial memory strategy by selecting appropriate samples for memorizing.We compare the performance of our proposed method with three baselines through a large number of simulation experiments.Results show that our method achieves far higher classification accuracy than finetuning method and feature extraction method.Furthermore,it performs closely to joint training method which uses all old data in terms of classification accuracy which validates the effectiveness of our method against catastrophic forgetting.展开更多
基金supported by National Natural Science Foundation of China(No.51208053)the Special Fund for Basic Scientific Research of Central Colleges,Chang'an University(No.00092014G1211011,2013G1211009)
文摘To accurately assess the safety of freeway work zones, this paper investigates the safety of vehicle lane change maneuvers with improved cellular automata model. Taking the traffic conflict and standard deviation of operating speed as the evaluation indexes, the study evaluates the freeway work zone safety. With improved deceleration probability in car-following rules and the addition of lanechanging rules under critical state, the lane-changing behavior under critical state is defined as a conflict count. Through 72 schemes of simulation runs, the possible states of the traffic flow are carefully studied. The results show that under the condition of constant saturation traffic conflict count and vehicle speed standard deviation reach their maximums when the mixed rate of heave vehicles is 40%. Meanwhile, in the case of constant heavy vehicles mix, traffic conflict count and vehicle speed standard deviation reach maximum values when saturation rate is 0.75. Integrating all simulation results, it is known the traffic safety in freeway work zones is classified into four levels： safe, relatively safe, relatively dangerous, and dangerous.
文摘Radio modulation classification has always been an important technology in the field of communications.The difficulty of incremental learning in radio modulation classification is that learning new tasks will lead to catastrophic forgetting of old tasks.In this paper,we propose a sample memory and recall framework for incremental learning of radio modulation classification.For data with different signal-to-noise ratios,we use a partial memory strategy by selecting appropriate samples for memorizing.We compare the performance of our proposed method with three baselines through a large number of simulation experiments.Results show that our method achieves far higher classification accuracy than finetuning method and feature extraction method.Furthermore,it performs closely to joint training method which uses all old data in terms of classification accuracy which validates the effectiveness of our method against catastrophic forgetting.
