Robotic environmental monitoring: challenges, opportunities, and novel advancements

an IROS 2021 Event

September 27- October 01, 2021 - Prague, Czech Republic, Online


Global warming and pollution are threatening the survival of one million over the eight million species in forests and oceans on the planet. Continuous monitoring of the ecosystems is of paramount importance to protect and preserve this biodiversity. Today, human operators are the main option to perform these monitoring tasks. This is not only because of their specific expertise and knowledge in classifying plants and habitats but also because of their physical intelligence allowing them to move for hours in wild unstructured environments such as dunes, forests, and mountains.

Robotics has the possibility to enhance environmental preservation through the active assistance of the operators, lightening the burden on the human workforce. However, the challenges introduced by the deployment of robots in real natural environments are numerous. Therefore, this goal requires a capitalization of the newest robotic technologies and methodologies going through several different research fields. Autonomous classification of plant species and monitoring of natural habitats are the main objectives. In this context, AI algorithms for recognition and classification play a major role. However, this should be supported by robots able to acquire data autonomously, thus able to autonomously navigate in outdoor natural environments. Rough and irregular terrains, obstacles, slippage, sinking are just a few examples of the issues which need to be dealt with. This factor can be facilitated by proposing novel design and mechatronics for the robotic structure, also in conjunction with ad-hoc control architectures. In unstructured environments such as forests and dunes, interactions will be unavoidable, so robots should present an adaptive behavior and should be robust to potential unexpected external disturbances. In this context, the efficient exploitation of robot dynamics is crucial to allow long-lasting operations.

The primary goal of this full-day workshop is to critically discuss the current and new approaches used to tackle environmental monitoring. We invited speakers from communities such as ecology, conservation biology, botany, and biodiversity to firstly identify the main challenges and objectives posed by the monitoring of different habitats. Speakers from the robotic and AI community will then present current and future robotic solutions to deal with these demanding scenarios. A fundamental point in this workshop will be the applicability of the proposed solutions to real-world missions. For this reason, we will also discuss the technology readiness level (TRL) of the developed robots, and we will present a potential benchmarking approach for robots operating in natural unstructured environments.

Keywords

• Challenges of deploying robots in real natural environments

• Robotic platforms moving on irregular and complex natural terrains

• Motion planning in rough terrains, including e.g. gait generation, foot placement etc..

• Control algorithms for efficient, dynamic, and interacting motions

• Long-lasting autonomous outdoor navigation

• Mechatronic design of resilient, efficient, and adaptive robots

• Computer vision and perception

AI for classification and recognition of plants and natural habitats

• Benchmarking and standardization of operations in natural environments

• Sensor networks

A selection of experts in the field will alternate pre-recorded talks, organized in two different sessions: (1) insights from ecological research community and challenges presented by environmental monitoring; (2) theoretical and technological solutions to achieve an effective robotic environmental monitoring. An interactive discussion will be held on Oct 1, 14:00–16:00 CEST on the IROS21 platform. During this live event we will analyze the identified challenges, the proposed solutions, and the take-home messages.

This workshop is supported by the European Union’s Horizon 2020 research and innovation programme under grant agreement No. 101016970 (Natural Intelligence). The content of this publication is the sole responsibility of the authors. The European Commission or its services cannot be held responsible for any use that may be made of the information it contains.