TAE: A Semi-supervised Controllable Behavior-aware Trajectory Generator and Predictor
TAE: A Semi-supervised Controllable Behavior-aware Trajectory Generator and Predictor
复制标题
DOI：
10.1109/iros47612.2022.9981029
复制
发表时间：
2022-03
期刊：
影响因子：
--
通讯作者：
Ruochen Jiao;Xiangguo Liu;Bowen Zheng;Davis Liang;Qi Zhu
中科院分区：
文献类型：
--
作者：
Ruochen Jiao;Xiangguo Liu;Bowen Zheng;Davis Liang;Qi Zhu
轨迹的产生和预测是两项跨编织的任务，它们在智能车辆的计划者评估和决策中起着重要作用。大多数现有方法都集中在两者中之一，并经过优化，可以直接输出最终生成/预测的轨迹，该轨迹仅包含有限的信息，以增加关键场景增强和安全计划。在这项工作中，我们提出了一种新型的行为感知的轨迹自动编码器（TAE），该自动编码器（TAE）使用半监督的对抗自动编码器和运输中的域知识来明确地对驱动因素的行为（例如侵略性和意图）进行建模。我们的模型解决了统一体系结构和好处的轨迹产生和预测。这两个任务：该模型可以生成多样化，可控制和现实的轨迹，以增强计划者在安全至关重要和长尾的场景中进行操作，并可以提供关键行为的预测除了决策的最终轨迹。实验结果表明，我们的方法在轨迹产生和预测方面都可以实现有希望的表现。
Trajectory generation and prediction are two in-terwoven tasks that play important roles in planner evaluation and decision making for intelligent vehicles. Most existing methods focus on one of the two and are optimized to directly output the final generated/predicted trajectories, which only contain limited information for critical scenario augmentation and safe planning. In this work, we propose a novel behavior-aware Trajectory Autoencoder (TAE) that explicitly models drivers' behavior such as aggressiveness and intention in the latent space, using semi-supervised adversarial autoencoder and domain knowledge in transportation. Our model addresses trajectory generation and prediction in a unified architecture and benefits both tasks: the model can generate diverse, controllable and realistic trajectories to enhance planner op-timization in safety-critical and long-tailed scenarios, and it can provide prediction of critical behavior in addition to the final trajectories for decision making. Experimental results demonstrate that our method achieves promising performance on both trajectory generation and prediction.