BIDEONENA – Artificial Intelligence Techniques to reduce energy consumption in mobile robotics

General description of the project

Industrial mobile platforms are robots designed for the automated transport of loads in production environments, playing a key role in optimising workflows within factories. However, one of their main limiting factors is energy autonomy, which is directly conditioned by battery life.

The energy consumption of these platforms is influenced by multiple variables related both to the dynamics of movement itself and to operating conditions. As a consequence, limited autonomy requires frequent recharging, interrupting assigned tasks and reducing the overall productivity of the system. In addition, repeated charging and discharging cycles accelerate battery degradation, progressively reducing their capacity and exacerbating the problem in the mid- and long-term.

In this context, it is necessary to address strategies that improve the energy efficiency of mobile platforms and maximise their operating time without compromising safety or functionality in dynamic industrial environments.

Objectives

The objective of this study is to improve the energy efficiency and operational autonomy of industrial mobile platforms by applying artificial intelligence techniques to navigation and motion planning processes. The project aims to reduce the energy consumption associated with the movement of these systems without requiring modifications to existing hardware, thus contributing to a sustainable improvement in productivity and overall efficiency in industrial environments.

To this end, the study aims to analyse and model the relationship between the dynamic behaviour of mobile platforms, navigation trajectories and energy consumption, in order to identify more efficient movement patterns and strategies. Based on this knowledge, the aim is to integrate energy criteria into route planning processes, allowing robots to select trajectories that minimise consumption while maintaining operational, safety and process quality requirements.

The project also aims to extend these energy efficiency criteria to dynamic scenarios, addressing real-time navigation in the presence of moving obstacles and changes in the environment, without compromising the system's responsiveness. Finally, the aim is to develop evaluation and validation mechanisms that enable systematic analysis of the impact of the proposed strategies on energy consumption, autonomy and overall system performance, facilitating analysis prior to application in real industrial environments.

Contribution of NAIR Center

NAIR Center has made a key contribution to the development of advanced route planning strategies aimed at optimising the energy consumption of industrial mobile platforms using artificial intelligence techniques. Its contribution has focused on effectively integrating predictive energy consumption models into classic navigation algorithms, allowing trajectory planning to take into account not only geometric criteria, but also the actual energy impact of each movement decision.

Within this framework, NAIR developed a predictive energy consumption model specifically for an industrial mobile platform, capable of estimating the expenditure associated with different basic movements, such as translations and rotations. The model made it possible to capture complex non-linear relationships between operational variables, such as load, speed, acceleration, distance travelled and turning angles, and energy consumption, providing an accurate estimate that can be adapted to different operating conditions.

This knowledge was integrated into the A* planning algorithm, redefining its cost function to incorporate energy efficiency criteria. Based on this reformulation, two complementary strategies were developed that allow trajectories to be evaluated not only in terms of distance, but also in terms of their energy cost. These approaches give the algorithm the ability to select routes that minimise consumption without compromising the operational requirements of the system.

Experimental evaluation in simulated environments based on real industrial scenarios showed substantial reductions in energy consumption compared to traditional planning approaches, resulting in significant increases in the operational autonomy of mobile platforms.

Partners

Funding

Project funded by the 2023 Interregional Innovation Projects Programme

More projects