We present a reward-predictive, model-based deep learning method featuring trajectory-constrained visual attention for local planning in visual navigation tasks. Our method learns to place visual attention at locations in latent image space which follow trajectories caused by vehicle control actions to enhance predictive accuracy during planning. The attention model is jointly optimized by the task-specific loss and an additional trajectory-constraint loss, allowing adaptability yet encouraging a regularized structure for improved generalization and reliability. Importantly, visual attention is applied in latent feature map space instead of raw image space to promote efficient planning. We validated our model in visual navigation tasks of planning low turbulence, collision-free trajectories in off-road settings and hill climbing with locking differentials in the presence of slippery terrain. Experiments involved randomized procedural generated simulation and real-world environments. We found our method improved generalization and learning efficiency when compared to no-attention and self-attention alternatives.
翻译:我们展示了一种有奖励的、基于模型的深层次学习方法,在视觉导航任务的地方规划中,以轨迹限制的视觉关注方式进行视觉关注; 我们的方法学会了将视觉关注置于潜伏的图像空间中,随着车辆控制行动所引发的轨迹提高预测准确度,从而在规划期间提高预测准确性; 关注模式因特定任务的损失和额外的轨迹限制损失而共同优化,允许适应性,但又鼓励一种正规化的结构来改进一般化和可靠性。 重要的是,视觉关注应用在潜伏地图空间,而不是原始的图像空间,以促进有效的规划。 我们验证了我们的视觉导航任务模式,即规划在路外环境中的低波动、无碰撞轨迹和山坡上,在滑坡的地形出现时有锁定差异。 实验涉及随机化程序产生的模拟和现实世界环境。 我们发现,与不注意和自我注意的替代方法相比,我们的方法提高了一般化和学习效率。