To plan safely in uncertain environments, agents must balance utility with
safety constraints. Safe planning problems can be modeled as a
chance-constrained partially observable Markov decision process (CC-POMDP) and
solutions often use expensive rollouts or heuristics to estimate the optimal
value and action-selection policy. This work introduces the ConstrainedZero
policy iteration algorithm that solves CC-POMDPs in belief space by learning
neural network approximations of the optimal value and policy with an
additional network head that estimates the failure probability given a belief.
This failure probability guides safe action selection during online Monte Carlo
tree search (MCTS). To avoid overemphasizing search based on the failure
estimates, we introduce $\Delta$-MCTS, which uses adaptive conformal inference
to update the failure threshold during planning. The approach is tested on a
safety-critical POMDP benchmark, an aircraft collision avoidance system, and
the sustainability problem of safe CO$_2$ storage. Results show that by
separating safety constraints from the objective we can achieve a target level
of safety without optimizing the balance between rewards and costs.

在不确定环境中安全规划时，智能体需在效用和安全限制之间进行平衡。本研究提出了基于信念空间的 ConstrainedZero 政策迭代算法，通过学习神经网络近似的最优值和策略，并引入了额外的网络头来估计置信度下的失败概率，以指导在线蒙特卡洛树搜索中的安全动作选择。通过使用自适应符合推理更新规划中的失败阈值，引入了 Δ-MCTS，以避免过分强调基于失败估计的搜索。该方法在一个安全关键的 POMDP 基准、飞机碰撞避免系统以及安全二氧化碳储存可持续性问题上进行了测试，结果表明，通过将安全限制与目标分离，可以在不优化回报和成本之间的平衡的情况下实现目标安全水平。