Method get_reward

mbd/envs/car2d.py:89–93

(self, q)

    @partial(jax.jit, static_argnums=(0,))
    def get_reward(self, q):
        reward = (
            1.0 - (jnp.clip(jnp.linalg.norm(q[:2] - self.xg[:2]), 0.0, 0.2) / 0.2) ** 2
        )
        return reward

    @partial(jax.jit, static_argnums=(0,))
    def eval_xref_logpd(self, xs):