SSAGCN: Social Soft Attention Graph Convolution Network for Pedestrian Trajectory Prediction

Pedestrian trajectory prediction is an important technique of autonomous driving. In order to accurately predict the reasonable future trajectory of pedestrians, it is inevitable to consider social interactions among pedestrians and the influence of surrounding scene simultaneously, which can fully represent the complex behavior information and ensure the rationality of predicted trajectories obeyed realistic rules. In this article, we propose one new prediction model named social soft attention graph convolution network (SSAGCN), which aims to simultaneously handle social interactions among pedestrians and scene interactions between pedestrians and environments. In detail, when modeling social interaction, we propose a new social soft attention function, which fully considers various interaction factors among pedestrians. Also, it can distinguish the influence of pedestrians around the agent based on different factors under various situations. For the scene interaction, we propose one new sequential scene sharing mechanism. The influence of the scene on one agent at each moment can be shared with other neighbors through social soft attention; therefore, the influence of the scene is expanded both in spatial and temporal dimensions. With the help of these improvements, we successfully obtain socially and physically acceptable predicted trajectories. The experiments on public available datasets prove the effectiveness of SSAGCN and have achieved state-of-the-art results. The project code is available at https://github.com/WW-Tong/ssagcn_for_path_prediction.

Robot Navigation Based on Predicting of Human Interaction and its Reproducible Evaluation in a Densely Crowded Environment

To achieve robot navigation in crowded environments having high densities of moving people, it is insufficient to simply consider humans as moving obstacles and avoid collisions with them. That is, the impact of an approaching robot on human movements must be considered as well. Moreover, various navigation methods have been tested in their own environments in the literature, which made them difficult to compare with one another. Thus, we propose an autonomous robot navigation method in densely crowded environments for data-based predictions of robot-human interactions, together with a reproducible experimental test under controlled conditions. Based on localized positional relationships with humans, this method extracts multiple alternative paths, which can implement either following or avoidance, and selects an optimal path based on time efficiency. Each path is selected using neural networks, and the various paths are evaluated by predicting the position after a given amount of time has elapsed. These positions are then used to calculate the time required to reach a certain target position to ensure that the optimal path can be determined. We trained the predictor using simulated data and conducted experiments using an actual mobile robot in an environment where humans were walking around. Using our proposed method, collisions were avoided more effectively than when conventional navigation methods were used, and navigation was achieved with good time efficiency, resulting in an overall reduction in interference with humans. Thus, the proposed method enables an effective navigation in a densely crowded environment, while collecting human-interaction experience for further improvement of its performance in the future.

HyP-DESPOT: A hybrid parallel algorithm for online planning under uncertainty

Robust planning under uncertainty is critical for robots in uncertain, dynamic environments, but incurs high computational cost. State-of-the-art online search algorithms, such as DESPOT, have vastly improved the computational efficiency of planning under uncertainty and made it a valuable tool for robotics in practice. This work takes one step further by leveraging both CPU and GPU parallelization in order to achieve real-time online planning performance for complex tasks with large state, action, and observation spaces. Specifically, Hybrid Parallel DESPOT (HyP-DESPOT) is a massively parallel online planning algorithm that integrates CPU and GPU parallelism in a multi-level scheme. It performs parallel DESPOT tree search by simultaneously traversing multiple independent paths using multi-core CPUs; it performs parallel Monte Carlo simulations at the leaf nodes of the search tree using GPUs. HyP-DESPOT provably converges in finite time under moderate conditions and guarantees near-optimality of the solution. Experimental results show that HyP-DESPOT speeds up online planning by up to a factor of several hundred in several challenging robotic tasks in simulation, compared with the original DESPOT algorithm. It also exhibits real-time performance on a robot vehicle navigating among many pedestrians.