The Safeforest project, a CMU Portugal Large Scale Collaborative Project, hosted a final Workshop on May 4th at the Instituto de Sistemas e Robótica da Universidade de Coimbra. The event brought together members of the consortia and a diverse audience to showcase the project’s closing remarks and successful results.
Funded through PT2020 and FCT under the CMU Portugal Program, the Safeforest – Semi-Autonomous Robotic System for Forest Cleaning and Fire Prevention is a groundbreaking initiative led by the company Ingeniarius, in partnership with SILVAPOR, ADAI (Associação para o Desenvolvimento da Aerodinâmica Industrial), ISR from Universidade de Coimbra and CMU (Carnegie Mellon University).
Its primary objective is to revolutionize forest wildfire prevention by developing an advanced robotic system for the prevention of wildland and wildland-urban interface fires. The robotic system will focus on creating wide protection corridors used as fuel barriers to control and reduce the spread of large forest fires.
The Opening session was kicked-off by Prof. Aníbal Traça de Almeida, providing an insightful introduction to the project and its noteworthy outcomes. This was followed by a presentation from Sílvia Castro, CMU Portugal Executive Director, who shed light on the Program’s overall activities, placing special emphasis on the impactful Large Scale Projects initiative, which was launched in 2018 and is set to conclude by June 2023.
The morning was dedicated to a technical presentation showcasing the remarkable advancements made by the Safeproject consortium in the field of robotic systems. Esteemed researchers from the collaborating entities took the stage to share their contributions: Micael Couceiro (Ingeniarius); Beril Yalcinkaya (Ingeniarius), Carlos Viegas (ADAI), Paulo Peixoto (ISR), Luís Conde (ISR), David Russell (CMU).
The next panel delved on societal topics related to the importance of forest management as a measure to promote their resilience. Virgílio Bernardino from Silvapor and Miguel Almeida from ADAI, led the discussion presenting the research done on the characterization of houses and surroundings in rural areas in Portugal, protection measures to prevent wildfires, and other related topics.
In the afternoon, the workshop shifted gears to outdoor demonstrations, showcasing the operational capabilities of two key robots: the SCOUT drone and the semi-autonomous platform RANGER.
George Kantor (PI of the Project at CMU Robotics Institute), Maria Eduarda Andrada (CMU Affiliated Ph.D. student at Universidade de Coimbra) and David Russell (M.Sc. student at CMU), traveled from Pittsburgh for the event, one of multiple trips during this three year project. “The SafeForest project provided a fantastic opportunity for my group at CMU to apply our research in a new domain (forestry) with new partners at universities and companies in Portugal. We were able to adapt a sensing payload that had previously been used from terrestrial vehicles in agricultural settings for us on aerial drones for mapping forests. The core CMU team was composed of two professors, a project scientist, and two masters students. Through multiple trips both ways across the Atlantic, we all had plenty of opportunities to interact directly with Portuguese team members in the US and in Portugal “, shared George Kantor.
The CMU team then showcased the demonstration of SCOUT’s capacities to perform the mapping of the terrain and the identification of the forest areas to be cleared.
Moving to the RANGER, this semi-autonomous mobile platform consists of a compact track loader with a sensor toolkit installed, capable of removing excess vegetation based on the preliminary mapping of the area, achieving the necessary clearing of fuel breaks.
The integration of this comprehensive system allows for terrain management planning using the map data gathered by the SCOUT, needed to effectively plan missions and trajectories for the RANGER, optimizing time and cost-effectiveness. The system presents these plans to the user, offering the opportunity to review, simulate, and make necessary adjustments. Designed with a user-friendly interface and intuitive operation, the system caters to both forest operators and landowners.
According to Micael Couceiro, Ingeniarius CEO, “the development of these two semi-autonomous robots has significantly enhanced land clearing missions compared to traditional methods. The software designed for terrain and vegetation mapping, along with the platform for managing and visualizing the collected data, have been instrumental in enabling more effective clearing missions.”
By employing these mobile robots, it is possible to remove redundant vegetation efficiently and safely, reducing the risk of forest wildfires. Moreover, the ability for robots to navigate semi-autonomously reduces hazards for human operators, minimizes the environmental impact of land clearing by reducing, for example, the use of herbicides, and improves the adaptability to different types of terrain and weather conditions.
However, as noted by Micael Couceiro “While the Safeforest project has achieved significant results, there is still further work required to make the solution commercially viable. In particular, we need to improve the integration between perception and action of robots, enabling them to operate more effectively in complex and dynamic environments, such as those found in forests.”
The PI at CMU added “there is still much work to do, but the SafeForest project represents an important and tangible step toward using robotics and AI to mitigate potential wild fires around the world. ”
The event also counted with the presence of the 2nd Regional Commander Francisco Peraboa, representing the Autoridade Nacional de Emergência e Proteção Civil (the National Emergency and Civil Protection Authority).
The three year project which started in 2019, will end by June 2023. However, the team has plans to further develop these technologies in the future by applying to other funding initiatives.
The Safeforest Project – Semi-Autonomous Robotic System for Forest Cleaning and Fire Prevention (CENTRO-01-0247-FEDER-045931) – is co-financed by FEDER, COMPETE 2020, by the CENTRO2020 Program and by Fundação para a Ciência e Tecnologia (FCT) under the CMU Portugal Program.
More about the project here.