Reinforced iLQR: A Sample-Efficient Robot Locomotion Learning | IEEE Conference Publication | IEEE Xplore