Notably, a reward function was designed considering contact points and phases, which enforced gait symmetry and periodicity, culminating in an improved bounding performance. The DRL methods ...