Nowadays robots are able to perform decisions independent from the operator. Also, they are able to form a various kind of unions, such as swarms, schools, or coalitions to perform joint task solving. The most powerful and flexible type of union is coalitions. Due to each robot acts like an independent agent it is important to provide trusted interaction between them. It is quite hard to do this with existing methods based only on the knowledge representation using ontologies and reasoning techniques. The paper proposes to use smart contracts in blockchain to enrich the knowledge-based system by functions, specific for blockchain such as immutable transaction log, consensus between all participants as well as possibility to automate control on task resolving. The paper describes the new environment framework based on integration of cyberphysical system and blockchain, and interaction model between all framework elements using BPMN 2.0 notation. Smart contracts presented in the paper provide functions for tasks distribution between robots, resource allocation, and monitoring the task execution and reward distribution.