【正文】 on actuated traffic controllers and is able to pro actively handle traffic situations and 3 handling the different, sometimes conflicting, aims of traffic controllers. The proposed use of the concept of agents in this research is experimental. Assumptions and considerations on agent based urban traffic control There are three aspects where agent based traffic control and management can improve current state of the art UTC systems: Adaptability. Intelligent agents are able to adapt its behaviour and can learn from earlier situations. Communication. Communication makes it possible for agents to cooperate and tune signal plans. Proactive behaviour. Due to the pro active behaviour traffic control systems are able to plan ahead. To be acceptable as replacement unit for current traffic control units, the system should perform the same or better than current systems. The agent based UTC will require online and proactive reaction on changing traffic patterns. An agent based UTC should be demand responsive as well as adaptive during all stages and times. New methods for traffic control and traffic prediction should be developed as current ones do not suffice and cannot be used in agent technology. The adaptability can also be divided in several different time scales where the system may need to handle in a different way (Rogier, 1999): gradual changes due to changing traffic volumes over a longer period of time, abrupt changes due to changing traffic volumes over a longer period of time, abrupt, temporal, changes due to changing traffic volumes over a short period of time, abrupt, temporal, changes due to prioritised traffic over a short period of time One way of handling the balance between performance and plexity is the use of a hierarchical system layout. We propose a hierarchy of agents where every agent is responsible for its own optimal solution, but may not only be influenced by adjoining agents but also via higher level agents. These agents have the task of solving conflicts between lower level agents that they can39。s. One final aspect to be mentioned is the robustness of agent based systems (if all munication fails the agent runs on, if the agent fails a fixed program can be executed. To be able to keep our first urban traffic control model as simple as possible we have made the following assumptions: we limit ourselves to inner city traffic control (road segments, intersections, corridors), we handle only controlled intersections with detectors (intensity and speed) at all road segments, we only handle cars and we use simple rule bases for knowledge representation. Types of agents in urban intersection control As we divide the system in several, recognisable, parts we define the following 4 types of agents: Roads are represented by special road segment agents (RSA), Controlled intersections are represented by intersection agents (ITSA), 4 For specific, defined, areas there is an area agent (higher level), For specific routes there can be route agents, that spans several adjoining road segments (higher level). We have not chosen for one agent per signal. This may result in a more simple solution but available traffic control programs do not fit in that kind of agent. We deliberately choose a more plex agent to be able to use standard traffic control design algorithms and programs. The idea still is the optimisation on a local level (intersection), but with local and global control. Therefor we use area agents and route agents. All munication takes place between neighbouring agents and upper and lower level ones. Design of our agent based system The essence of a, demand responsive and proactive agent based UTC consists of several ITSA39。 from / to other ITSA39。 analysis (with an accurate model of the surrounds and knowing the traffic and traffic control rules define current trend。 calculation (calculate the next, optimal, cycle mathematically correct)。 handle current traffic problems)。s. The ITSA is the agent that controls and operates one specific intersection of which it is pletely informed. All ITSA39。s, RSA39。s and partly placed in the ITSA39。 Kim, 1994). Control strategy model The prediction of the prediction model is used in the control strategy planning phase. We have also included a performance indicating agent, necessary to update the control parameters in the slower loop. The control strategy agent uses the estimates of the prediction model agent to calculate the most optimal control strategy to proact on 6 the forecasts of the prediction model agent, checks with other adjoining agents its proposed traffic control schema and then plans the signal control strategy The munication schema is based on direct agent to agent munication via a work l