Modeling Pedestrian Motion in Crowded Scenes Based on the Shortest Path Principle

In the computer vision field, understanding human dynamics is not only a great challenge but also very meaningful work, which plays an indispensable role in public safety. Despite the complexity of human dynamics, physicists have found that pedestrian motion in a crowd is governed by some internal rules, which can be formulated as a motion model, and an effective model is of great importance for understanding and reconstructing human dynamics in various scenes. In this paper, we revisit the related research in social psychology and propose a two-part motion model based on the shortest path principle. One part of the model seeks the origin and destination of a pedestrian, and the other part generates the movement path of the pedestrian. With the proposed motion model, we simulated the movement behavior of pedestrians and classified them into various patterns. We next reconstructed the crowd motions in a real-world scene. In addition, to evaluate the effectiveness of the model in crowd motion simulations, we created a new indicator to quantitatively measure the correlation between two groups of crowd motion trajectories. The experimental results show that our motion model outperformed the state-of-the-art model in the above applications.

The Improved Motion Model of Particles in the Cone Crusher Considering the Spatial Compound Motion of the Mantle

The cone crusher is the main equipment in the particle crushing process. The productivity of the cone crusher is determined by the motion characteristics of particles passing through the crushing chamber. In order to accurately describe the motion characteristics of the particles, the influence of the spatial compound motion of the mantle rotates around the central axis of the cone crusher and its central axis on the motion characteristics of the particles is investigated, then the improved motion model is established. Through the coordinate system transformation matrix, the motion characteristics of the particles including spatial sliding, free-falling, and spatial compound falling are solved. The applicability and accuracy of the improved model in describing the motion characteristics of the particle were verified through the experiment using a reduced-scale experimental cone crusher to simulate the motion characteristics of the particle. Based on the improved model, the motion characteristics of the particles in the CF11 hydraulic cone crusher can be simulated. With the decrease in height, the motion characteristics of particles gradually change from spatial sliding to spatial compound falling and finally to free-falling. The particles deflect circumferentially around the central axis of the cone crusher. The circumferential deflection of particles is directly related to the motion characteristics including spatial sliding and spatial compound falling. The improved model provides a theoretical basis for the high energy design of the crushing chamber and productivity improvement of the cone crusher.

A Far-Field Ground-Motion Model for the North Australian Craton from Plate-Margin Earthquakes

ABSTRACT The Australian territory is just over 400 km from an active convergent plate margin with the collision of the Sunda–Banda Arc with the Precambrian and Palaeozoic Australian continental crust. Seismic energy from earthquakes in the northern Australian plate-margin region are channeled efficiently through the low-attenuation North Australian craton (NAC), with moderate-sized (Mw≥5.0) earthquakes in the Banda Sea commonly felt in northern Australia. A far-field ground-motion model (GMM) has been developed for use in seismic hazard studies for sites located within the NAC. The model is applicable for hypocentral distances of approximately 500–1500 km and magnitudes up to Mw 8.0. The GMM provides coefficients for peak ground acceleration, peak ground velocity, and 5%-damped pseudospectral acceleration at 20 oscillator periods from 0.1 to 10 s. A strong hypocentral depth dependence is observed in empirical data, with earthquakes occurring at depths of 100–200 km demonstrating larger amplitudes for short-period ground motions than events with shallower hypocenters. The depth dependence of ground motion diminishes with longer spectral periods, suggesting that the relatively larger ground motions for deeper earthquake hypocenters may be due to more compact ruptures producing higher stress drops at depth. Compared with the mean Next Generation Attenuation-East GMM developed for the central and eastern United States (which is applicable for a similar distance range), the NAC GMM demonstrates significantly higher short-period ground motion for Banda Sea events, transitioning to lower relative accelerations for longer period ground motions.

Automatic Vehicle Route Prediction based on Multi-Sensor Fusion

To solve the problem of inaccurate results of vehicle routing prediction caused by a large number of uncertain information collected by different sensors in previous automatic vehicle route prediction algorithms, an automatic vehicle route prediction algorithm based on multi-sensor fusion is studied. The process of fusion of multi-sensor information based on the D-S evidence reasoning fusion algorithm is applied to automatic vehicle route prediction. According to the contribution of a longitudinal acceleration sensor and yaw angular velocity sensor detection information to the corresponding motion model, the basic probability assignment function of each vehicle motion model is obtained; the basic probability assignment function of each motion model is synthesized by using D-S evidence reasoning synthesis formula. The new probability allocation of each motion model is obtained under all evidence and then deduced according to the decision rules. Guided by the current optimal motion model, the optimal motion model at each time is used to accurately predict the vehicle movement route. The simulation results show that the prediction error of the algorithm is less than 4% in the process of 30 minutes of automatic vehicle route prediction.

Autonomous maneuver strategy of swarm air combat based on DDPG

Unmanned aerial vehicles (UAVs) have been found significantly important in the air combats, where intelligent and swarms of UAVs will be able to tackle with the tasks of high complexity and dynamics. The key to empower the UAVs with such capability is the autonomous maneuver decision making. In this paper, an autonomous maneuver strategy of UAV swarms in beyond visual range air combat based on reinforcement learning is proposed. First, based on the process of air combat and the constraints of the swarm, the motion model of UAV and the multi-to-one air combat model are established. Second, a two-stage maneuver strategy based on air combat principles is designed which include inter-vehicle collaboration and target-vehicle confrontation. Then, a swarm air combat algorithm based on deep deterministic policy gradient strategy (DDPG) is proposed for online strategy training. Finally, the effectiveness of the proposed algorithm is validated by multi-scene simulations. The results show that the algorithm is suitable for UAV swarms of different scales.