Search

Trajectory prediction of multiple agents in a crowded scene is an essential component in many applications, including intelligent monitoring, autonomous robotics, and self-driving cars. Accurate agent trajectory prediction remains a significant challenge because of the complex dynamic interactions among the agents and between them and the surrounding scene. To address the challenge, we propose a decoupled attention-based spatial-temporal modeling strategy in the proposed trajectory prediction method. The past and current interactions among agents are dynamically and adaptively summarized by two separate attention-based networks and have proven powerful in improving the prediction accuracy. Moreover, it is optional in the proposed method to make use of the road map and the plan of the ego-agent for scene-compliant and accurate predictions. The road map feature is efficiently extracted by a convolutional neural network, and the features of the ego-agent’s plan is extracted by a gated recurrent network with an attention module based on the temporal characteristic. Experiments on benchmark trajectory prediction datasets demonstrate that the proposed method is effective when the ego-agent plan and the the surrounding scene information are provided and achieves state-of-the-art performance with only the observed trajectories.

Knowledge of vertical air motion in the atmosphere is important for both meteorological and climate studies due to its impact on clouds, precipitation and the vertical transport of air masses, heat, momentum, and composition. The vertical velocity (VV) of air is among the most difficult and uncertain quantities to measure due to its generally small magnitude and high temporal and spatial variability. In this study, a descending radiosonde system is developed to derive VV at the low and middle troposphere in north China during the summer months. The VV is estimated from the difference between the observed radiosonde descent speed and the calculated radiosonde descent speed in still air based on the fluid dynamic principle. The results showed that the estimated VV generally ranged from &minus, 1 m/s to 1 m/s, accounting for 80.2% of data points. In convective conditions, a wider distribution of the VV was observed, which was skewed to large values relative to those in nonconvective conditions. The average VV throughout the entire profile was close to 0 m/s under nonconvective conditions. In contrast, distinctive vertical air motions below 5 km above the ground were recorded under convective activities. Vigorous air motions with an absolute VV &gt, 2 m/s were occasionally observed and were often associated with the occurrence of cloud layers. Moreover, the detailed structure of the instant air motion near the cloud boundaries (i.e., top and base), with an absolute VV &gt, 10 m/s in convective weather systems, was clearly revealed by this technique. The uncertainty estimation indicated that this method has the potential to capture and describe events with vertical air motions &gt, 0.69 m/s, which is useful for a convective weather study. Further studies are required to carefully assess the accuracy and precision of this novel VV estimation technique.

Pedestrian trajectory prediction is crucial for many important applications. This problem is a great challenge because of complicated interactions among pedestrians. Previous methods model only the pairwise interactions between pedestrians, which not only oversimplifies the interactions among pedestrians but also is computationally inefficient. In this paper, we propose a novel model StarNet to deal with these issues. StarNet has a star topology which includes a unique hub network and multiple host networks. The hub network takes observed trajectories of all pedestrians to produce a comprehensive description of the interpersonal interactions. Then the host networks, each of which corresponds to one pedestrian, consult the description and predict future trajectories. The star topology gives StarNet two advantages over conventional models. First, StarNet is able to consider the collective influence among all pedestrians in the hub network, making more accurate predictions. Second, StarNet is computationally efficient since the number of host network is linear to the number of pedestrians. Experiments on multiple public datasets demonstrate that StarNet outperforms multiple state-of-the-arts by a large margin in terms of both accuracy and efficiency., Comment: Accepted by the 2019 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2019)

A common practice for autonomous driving is to train a model to mimic expert actions. However, the actions are randomly drawn based on a underlying policy. The policy specifies the probability distribution of the expert taking different actions under each circumstance. An ideal driving model should mimic the policy, rather than the action. In this paper, we propose a new method to mimic the policy. Our method employs deep neural network to map each circumstance to the probability distribution of actions. In practice, the distribution is approximately represented by its low order moments. Hence, our model is called Moments Deep Neural Network (MDNN). To improve the efficiency, MDNN employs a new type of dilated convolution we term the Dilated Tensor Convolution (DTC). DTC applies the idea of dilated convolution on every dimension of the input, reducing the number of parameters in MDNN. On multiple experiments, MDNN is demonstrated to outperform the classical deep neural network PilotNet using much less parameters.