Embodied Explainable AI: Interpretability Frameworks for Humanoid Robot Decision-Making and Kinematic Recourse

Embodied XAI is an absolute prerequisite for the safe, transparent deployment of humanoid robots in human-centric spaces. Transforming black-box neural controllers into legible physical agents requires a tight coupling of real-time post-hoc attribution, natural language explanations, and expressive kinematic design. Future research must focus on optimizing E-XAI algorithms to run efficiently within sub-millisecond robotic control loops without compromising physical safety or explanation accuracy.

[1] O. Khatib, "Real-time obstacle avoidance for manipulators and mobile robots," Int. J. Robotics Res., vol. 5, no. 1, pp. 90–98, 1986. [2] A. Dragan, K. Lee, and S. Srinivasa, "Legibility and predictability of robot motion," Proc. ACM/IEEE Int. Conf. Human-Robot Interaction (HRI), 2013, pp. 301–308. [3] M. T. Ribeiro, S. Singh, and C. Guestrin, "'Why should I trust you?': Explaining the predictions of any classifier," in Proc. ACM SIGKDD Int. Conf. Knowledge Discovery and Data Mining, 2016, pp. 1135–1144. [4] R. Guidotti, A. Monreale, S. Ruggieri, F. Turini, F. Giannotti, and D. Pedreschi, "A survey of methods for explaining black box models," ACM Comput. Surv., vol. 51, no. 5, pp. 1–42, 2018. [5] C. Molnar, Interpretable Machine Learning: A Guide for Making Black Box Models Explainable, 2nd ed. Munich, Germany: Leanpub, 2022. [6] S. M. Lundberg and S.-I. Lee, "A unified approach to interpreting model predictions," in Advances in Neural Information Processing Systems (NeurIPS), 2017, pp. 4765–4774. [7] S. Anca et al., "Modelling user expectations for human-robot interaction," Nature Machine Intelligence, vol. 2, no. 8, pp. 442–450, 2020. [8] S. Wachter, B. Mittelstadt, and C. Russell, "Counterfactual explanations without opening the black box: Automated decisions and the GDPR," Harvard Journal of Law & Technology, vol. 31, p. 841, 2018. [9] B. Kim et al., "Interpretability beyond feature attribution: Quantitative testing with concept activation vectors (TCAV)," in Proc. Int. Conf. Machine Learning (ICML), 2018, pp. 2668–2677. [10] D. Slack, S. Hilgard, E. Jia, S. Singh, and H. Lakkaraju, "Fooling LIME and SHAP: Adversarial attacks on post hoc explanation methods," in Proc. AAAI Conf. Human Computation and Crowdsourcing, 2020, pp. 180–186.