AATEAM: Achieving the Ad Hoc Teamwork by Employing the Attention Mechanism

In the ad hoc teamwork setting, a team of agents needs to perform a task without prior coordination. The most advanced approach learns policies based on previous experiences and reuses one of the policies to interact with new teammates. However, the selected policy in many cases is sub-optimal. Switching between policies to adapt to new teammates' behaviour takes time, which threatens the successful performance of a task. In this paper, we propose AATEAM – a method that uses the attention-based neural networks to cope with new teammates' behaviour in real-time. We train one attention network per teammate type. The attention networks learn both to extract the temporal correlations from the sequence of states (i.e. contexts) and the mapping from contexts to actions. Each attention network also learns to predict a future state given the current context and its output action. The prediction accuracies help to determine which actions the ad hoc agent should take. We perform extensive experiments to show the effectiveness of our method.

Multisensory System for Monitoring and Control of Artificial Plant Growing Environment

The paper represents a small controller with wireless connection, which measures air temperature, air humidity and soil humidity. As an output action it can switch on and off an irrigation pump. Monitoring is realized over Wi-Fi connection to a gateway and cloud service. The service allows database storage and graph representation of the input values via desktop or mobile device browser. Sensors choice is also explained in details with advantages and disadvantages of the possible solutions. Developed device allows fine tuning, according to the target plants, can be duplicated many times in the range of the hothouse. It is one of the cheapest solutions which provides very stable work as a result of the specific implementation.