Safe Reinforcement Learning and Adaptive Optimal Control With Applications to Obstacle Avoidance Problem | Synapse