SyRoC: Symbiotic robotics for QoS-aware heterogeneous applications in IoT-edge-cloud computing paradigm

Smart Robots, as an advanced domain of widespread concern, can be applied in diverse fields to perform compute-intensive Internet of Things (IoT) applications. However, their onboard resource constraints necessitate using the computation offloading technique to extend their capabilities. Additionally, the distributed locations of robots and the diversified quality of service (QoS) requirements of different applications pose challenges to the existing offloading works. To address these concerns, we put forward a collaborative offloading scheme for QoS-aware utility optimization for multi-robot system, and the optimization problem is formulated based on multi-attribute utility theory. Our methodology consists of three critical components: (1) we construct a symbiotic collaboration model that leverages the advantages of multiple computation modes across IoT-edge-cloud infrastructure and tap the potential of multi-hop relay transmission technology to enable efficient collaborative offloading; (2) we develop a QoS-aware preference assessment framework based on analytic hierarchy process to effectively learn the preference weights of various applications executed and prioritizations for key QoS parameters affecting strategy-making of robots; and (3) we apply the game-theoretic approach to devise a symbiotic robotics computation offloading algorithm named SyRoC, which operates in a distributed manner to solve the optimization problem and determine the most effective strategies. To validate our scheme, we conduct a series of experiments on the software simulator, which demonstrate that our proposed scheme not only realizes stable performance on guaranteeing convergence and system utility when robots move dynamically, but also outperforms five typical algorithms in key metrics such as energy consumption (13.43% 4,), delay (14.22% 4,), and average QoS utility of robots corresponding to the four types of applications executed (12.80% f, 7.30% f, 7.90% f, and 12.71% f) in scenarios involving the expansion of robot scales.(c) 2023 Elsevier B.V. All rights reserved.