Relative Velocity-Based Reward Model for Socially-Aware Navigation with Deep Reinforcement Learning | IEEE Conference Publication | IEEE Xplore