arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2207.06982eess.SYcs.SY
收藏

基于模型控制的对抗性实例:敏感性分析

Adversarial Examples for Model-Based Control: A Sensitivity Analysis

Po-han Li, Ufuk Topcu, Sandeep P. Chinchali

摘要:

我们提出了一种方法来攻击依赖于外部时间序列预测作为任务参数的控制器。对手可以通过伪造时间序列来操纵控制器的成本、状态和动作,在这种情况下扰乱了真实时间序列。由于控制器通常在其成本和约束中编码安全要求或能量限制,因此我们将这种操纵称为对抗性攻击。我们表明,不同的攻击基于模型的控制器可以增加控制成本,激活约束,甚至使控制优化问题不可行。我们使用线性二次型调节器和凸模型预测控制器作为对抗性攻击如何成功的例子,并展示了对抗性攻击对电网运营商电池存储控制任务的影响。因此,我们的方法增加了控制成本的$8500\%$和能量约束的$13\%$对真实的电力需求时间序列。

英文摘要:

We propose a method to attack controllers that rely on external timeseries forecasts as task parameters. An adversary can manipulate the costs, states, and actions of the controllers by forging the timeseries, in this case perturbing the real timeseries. Since the controllers often encode safety requirements or energy limits in their costs and constraints, we refer to such manipulation as an adversarial attack. We show that different attacks on model-based controllers can increase control costs, activate constraints, or even make the control optimization problem infeasible. We use the linear quadratic regulator and convex model predictive controllers as examples of how adversarial attacks succeed and demonstrate the impact of adversarial attacks on a battery storage control task for power grid operators. As a result, our method increases control cost by $8500\%$ and energy constraints by $13\%$ on real electricity demand timeseries.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑