Towards Probabilistic Causal Discovery, Inference & Explanations for Autonomous Drones in Mine Surveying Tasks (2308.10047v2)
Abstract: Causal modelling offers great potential to provide autonomous agents the ability to understand the data-generation process that governs their interactions with the world. Such models capture formal knowledge as well as probabilistic representations of noise and uncertainty typically encountered by autonomous robots in real-world environments. Thus, causality can aid autonomous agents in making decisions and explaining outcomes, but deploying causality in such a manner introduces new challenges. Here we identify challenges relating to causality in the context of a drone system operating in a salt mine. Such environments are challenging for autonomous agents because of the presence of confounders, non-stationarity, and a difficulty in building complete causal models ahead of time. To address these issues, we propose a probabilistic causal framework consisting of: causally-informed POMDP planning, online SCM adaptation, and post-hoc counterfactual explanations. Further, we outline planned experimentation to evaluate the framework integrated with a drone system in simulated mine environments and on a real-world mine dataset.
- R. Cannizzaro and L. Kunze, “CAR-DESPOT: Causally-informed online POMDP planning for robots in confounded environments,” 2023.
- T. Hellström, “The relevance of causation in robotics: a review, categorization, and analysis,” Paladyn, 2021.
- H. Kurniawati, “Partially observable markov decision processes and robotics,” Annual Review of Control, Robotics, and Autonomous Systems, vol. 5, no. 1, pp. 253–277, 2022.
- M. Diehl and K. Ramirez-Amaro, “A causal-based approach to explain, predict and prevent failures in robotic tasks,” 2022.
- E. Bareinboim, A. Forney, and J. Pearl, “Bandits with unobserved confounders: A causal approach,” in Advances in Neural Information Processing Systems. Curran Associates, Inc., 2015.
- M. Gasse, D. Grasset, G. Gaudron, and P.-Y. Oudeyer, “Causal reinforcement learning using observational and interventional data,” 2021.
- L. Castri, S. Mghames, and N. Bellotto, “From continual learning to causal discovery in robotics,” in Proceedings of The First AAAI Bridge Program on Continual Causality. PMLR, 2023.
- C. K. Assaad, E. Devijver, and E. Gaussier, “Survey and evaluation of causal discovery methods for time series,” JAIR, 2022.
- R. Howard and L. Kunze, “Evaluating temporal observation-based causal discovery techniques applied to road driver behaviour,” in Proc. of the 2nd Conf. on Causal Learning and Reasoning. JMLR, 2023.
- B. Huang, K. Zhang, M. Gong, and C. Glymour, “Causal discovery and forecasting in nonstationary environments with state-space models,” in Proc. of the 36th Int. Conf. on Machine Learning. PMLR, 2019.
- L. Kunze, O. Gunes, D. Hillier, M. Munks, H. Webb, P. Salvini, D. Omeiza, and M. Jirotka, “Towards explainable and trustworthy collaborative robots through embodied question answering,” 2022.
- T. Gerstenberg, “What would have happened? counterfactuals, hypotheticals and causal judgements,” Philosophical Transactions of the Royal Society B: Biological Sciences, vol. 377, 2022.
- T. Kloek and H. K. van Dijk, “Bayesian estimates of equation system parameters: An application of integration by monte carlo,” Econometrica, vol. 46, no. 1, pp. 1–19, 1978. [Online]. Available: http://www.jstor.org/stable/1913641