Yi Luo - Academia.edu (original) (raw)
Papers by Yi Luo
Proceedings of the 6th international joint conference on Autonomous agents and multiagent systems, 2007
Canonical problems are simplified representations of a class of real world problems. They allow r... more Canonical problems are simplified representations of a class of real world problems. They allow researchers to compare algorithms in a standard setting which captures the most important challenges of the real world problems being modeled. Such examples are the block world for planning, two-player games for algorithms which learn the behavior of the opponent agent, or the "split the pie" game for a large class of negotiation problems. In this paper we focus on negotiating collaboration in space and time, a problem with many important real world applications. Although technically a multi-issue negotiation, we show that the problem can not be represented in a satisfactory manner by the split the pie model. We propose the "children in the rectangular forest" (CRF) model as a possible canonical problem for negotiating spatio-temporal collaboration. By exploring a centralized and a peer-to-peer negotiation based solution, we demonstrate that the problem captures the main challenges of the real world problems while allows us to simplify away some of the computationally demanding but semantically marginal features of real world problems.
Studies in Computational Intelligence, 2010
In the convoy formation problem, two embodied agents are negotiating the synchronization of their... more In the convoy formation problem, two embodied agents are negotiating the synchronization of their movement for a portion of the path from their respective sources to destinations. We consider a setting where the negotiation happens in physical time, thus the agents have the opportunity to perform actions while negotiating. Thus, the agent’s behavior is controlled by the interacting pair of
Journal of Intelligent Transportation Systems, 2014
Current state-of-the-art highway traffic flow simulators rely extensively on models using formula... more Current state-of-the-art highway traffic flow simulators rely extensively on models using formulas similar to those describing physical phenomena, such as forces, viscosity, or potential fields. These models have been carefully calibrated to represent the overall flow of traffic and they can also be extended to account for the cognitive limitations of the driver, such as reaction times. However, there are some aspects of driver behavior, such as strategic planning, that are difficult to formulate mathematically. In this article, we describe the YAES-DSIM highway simulator, which integrates virtual physics models with an agent-based model. The virtual physics component models the physical vehicle and the subconscious aspects of the driver behavior, while the agent component is responsible for the strategic and tactical decisions, which are difficult to model using virtual physics. We focus on the lane change decisions of the drivers, with special attention to the optimal lane positioning for a safe exit. We have used the model to simulate the flow of traffic on Highway 408 in Orlando, Florida, and to study the impact of various tactical and strategic decisions on the efficiency and safety of the traffic.
Proceedings of the 6th international joint conference on Autonomous agents and multiagent systems, 2007
Canonical problems are simplified representations of a class of real world problems. They allow r... more Canonical problems are simplified representations of a class of real world problems. They allow researchers to compare algorithms in a standard setting which captures the most important challenges of the real world problems being modeled. Such examples are the block world for planning, two-player games for algorithms which learn the behavior of the opponent agent, or the "split the pie" game for a large class of negotiation problems. In this paper we focus on negotiating collaboration in space and time, a problem with many important real world applications. Although technically a multi-issue negotiation, we show that the problem can not be represented in a satisfactory manner by the split the pie model. We propose the "children in the rectangular forest" (CRF) model as a possible canonical problem for negotiating spatio-temporal collaboration. By exploring a centralized and a peer-to-peer negotiation based solution, we demonstrate that the problem captures the main challenges of the real world problems while allows us to simplify away some of the computationally demanding but semantically marginal features of real world problems.
Studies in Computational Intelligence, 2010
In the convoy formation problem, two embodied agents are negotiating the synchronization of their... more In the convoy formation problem, two embodied agents are negotiating the synchronization of their movement for a portion of the path from their respective sources to destinations. We consider a setting where the negotiation happens in physical time, thus the agents have the opportunity to perform actions while negotiating. Thus, the agent’s behavior is controlled by the interacting pair of
Journal of Intelligent Transportation Systems, 2014
Current state-of-the-art highway traffic flow simulators rely extensively on models using formula... more Current state-of-the-art highway traffic flow simulators rely extensively on models using formulas similar to those describing physical phenomena, such as forces, viscosity, or potential fields. These models have been carefully calibrated to represent the overall flow of traffic and they can also be extended to account for the cognitive limitations of the driver, such as reaction times. However, there are some aspects of driver behavior, such as strategic planning, that are difficult to formulate mathematically. In this article, we describe the YAES-DSIM highway simulator, which integrates virtual physics models with an agent-based model. The virtual physics component models the physical vehicle and the subconscious aspects of the driver behavior, while the agent component is responsible for the strategic and tactical decisions, which are difficult to model using virtual physics. We focus on the lane change decisions of the drivers, with special attention to the optimal lane positioning for a safe exit. We have used the model to simulate the flow of traffic on Highway 408 in Orlando, Florida, and to study the impact of various tactical and strategic decisions on the efficiency and safety of the traffic.