Joint Sensing Task Assignment and Collision-Free Trajectory Optimization for Mobile Vehicle Networks Using Mean-Field Games

Kang, Yuhan; Liu, Siting; Zhang, Hongliang; Li, Wuchen; Han, Zhu; Osher, Stanley; Poor, H. Vincent

doi:10.1109/jiot.2020.3047739

Cited by 21 publications

(5 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Multi-agent systems have attracted much attention in recent years [1][2][3]. As can be seen from previous research, navigation [4] and obstacle avoidance [5,6] deserve to be studied.…”

Section: Introductionmentioning

confidence: 99%

“…As can be seen from previous research, navigation [4] and obstacle avoidance [5,6] deserve to be studied. However, the existing avoidance mechanisms [3,[7][8][9] have at least one of the foundations, which are efficient interactions among agents, a clean environment, and full knowledge of the surroundings. In reality, particularly in military applications, attacks and external interference might disable the communication systems and simultaneous localization and mapping (SLAM) system [7,10].…”

Section: Introductionmentioning

confidence: 99%

“…After getting the velocity space, the problem that the contradicted constraints cause an invalid solution must be addressed, and we adopt a logic metric method [3,48]. By dividing concave VA-S into two convex parts and denoting them with binary variables, the search in the concave space can be finished by optimizing binary variables.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Collision and obstacle avoidance for agents based on velocity programming in unknown environments under kinematical constraints

Xiong

Luo

Liu

et al. 2023

IET Control Theory & Appl

View full text Add to dashboard Cite

This paper conducts research on obstacle and collision avoidance under kinematical constraints in unknown environments, while communication and simultaneous localization and mapping (SLAM) are unavailable to agents. Then a strategy based on mixed-integer programming is proposed, in which velocity constraints are established with a modified Barrier function for obstacles that can be completely detected. As for obstacles that cannot be completely detected, a feasible set is built for the velocity programming based on the convex theory, and the contradicted constraints are addressed with the logic metric method. Besides, the actuator saturation is avoided by converting kinematical constraints into the restrictions on the magnitude, the restrictions on the direction, and the negative correlations between the components of the velocity respectively. Given that invalid nominal velocity leads to collisions, a virtual goal is estimated for nominal velocity improvement. In addition, the local extremum brought by empty programming space due to multiple constraints is repaired by fixing the velocity constraints. The feasibility of the proposed strategy is analyzed, and numerical simulations are provided to verify the effectiveness of the proposed strategy.

show abstract

“…Multi-agent systems have attracted much attention in recent years [1][2][3]. As can be seen from previous research, navigation [4] and obstacle avoidance [5,6] deserve to be studied.…”

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Collision and obstacle avoidance for agents based on velocity programming in unknown environments under kinematical constraints

Xiong

Luo

Liu

et al. 2023

IET Control Theory & Appl

View full text Add to dashboard Cite

show abstract

“…The MFG framework, introduced by M. Huang, P. Caines, R. Malhamé [1, 2] and P.-L. Lions, J.-M. Lasry [3,4,5], is currently an active field with applications in economics [6,7,8,9], finance [10,11,12,7], industrial engineering [13,14,15], swarm robotics [16,17,18,19], epidemic modelling [20,21] and data science [22,23,24]. For comprehensive exposition of MFG systems we refer to [5,25,26] for nonlocal couplings, [27,28,26] for local couplings, [29,30] for a probabilistic approach, [31] for infinite-dimensional control approach, [32,33] for the master equation, and [34] for the control on the acceleration.…”

Section: Introductionmentioning

confidence: 99%

Random Features for High-Dimensional Nonlocal Mean-Field Games

Agrawal,

Lee,

Fung

et al. 2022

Preprint

View full text Add to dashboard Cite

We propose an efficient solution approach for high-dimensional nonlocal meanfield game (MFG) systems based on the Monte Carlo approximation of interaction kernels via random features. We avoid costly space-discretizations of interaction terms in the state-space by passing to the feature-space. This approach allows for a seamless mean-field extension of virtually any single-agent trajectory optimization algorithm. Here, we extend the direct transcription approach in optimal control to the mean-field setting. We demonstrate the efficiency of our method by solving MFG problems in high-dimensional spaces which were previously out of reach for conventional non-deep-learning techniques.

show abstract

“…Based on the derived outage probability, we formulate the AoI minimisation problem as a Stackelberg game as the IoT devices need to adjust their own strategies according the position of the UAV [15]. To accommodate a large number of uniform IoT devices, we propose a mean-field game based approach 1 [16]. To be specific, the behaviours of these IoT devices are approximated as a distribution.…”

Section: Introductionmentioning

confidence: 99%

Age‐optimal design for UAV‐assisted grant‐free non‐orthogonal massive access: Mean‐field game approach

Zhang

2022

IET Communications

Self Cite

View full text Add to dashboard Cite

Due to the capability of providing flexible network coverage, unmanned aerial vehicles (UAVs) are a promising solution to support the Internet‐of‐Things (IoT) applications, especially for the area without terrestrial network coverage. In this paper, a UAV‐assisted IoT network is considered, where a UAV serves as an aerial base station to serve ground IoT devices. To support the massive access, grant‐free non‐orthogonal communications are also adopted, where packets generated by IoT devices can skip the handshake process and share a channel with other packets for transmission. With the aim to maintain the information freshness, the age of information (AoI) for the network is first derived, and an AoI minimization problem is formulated as a Stackelberg game where the UAV is the leader and ground IoT devices are followers. However, solving such a problem is challenging as the number of involved followers is large. To simplify the problem, a mean‐field game based approach is proposed, where the average behaviours of followers are considered instead of individual strategies. Simulation results show that the computational complexity of the proposed scheme is almost unchanged with the number of IoT devices while the performance is close to the optimal one.

show abstract

Joint Sensing Task Assignment and Collision-Free Trajectory Optimization for Mobile Vehicle Networks Using Mean-Field Games

Cited by 21 publications

References 32 publications

Collision and obstacle avoidance for agents based on velocity programming in unknown environments under kinematical constraints

Collision and obstacle avoidance for agents based on velocity programming in unknown environments under kinematical constraints

Random Features for High-Dimensional Nonlocal Mean-Field Games

Age‐optimal design for UAV‐assisted grant‐free non‐orthogonal massive access: Mean‐field game approach

Contact Info

Product

Resources

About