Conflict in multi-agent systems is ubiquitous. Research often focuses on the process of resolving conflicts between different agents. We call this the inter-agent conflict resolution process. However, in complex problem solving agents the process of resolving conflicts with other agents impacts local problem solving as well as deals made with other agents. This leads to the need for an intra-agent conflict resolution process between the agent’s coordination mechanism and its local controller. In situations in which conflicts cannot be resolved to produce satisfactory solution paths for the agent, or a set of agents, it may be necessary to move the conflict resolution process to a higher level, a meta level, in which the agents negotiate to possibly revise their (individual and/or joint) high-level objectives. In this paper, we explore these different levels and their interdependence in the context of our research in multi-agent control and coordination.