Distributed Object Based SLAM
The use of multiple cooperative robots or mobile devices has the potential to enable fast information gathering, and more efficient coverage and monitoring of large areas. In particular, distributed SLAM, i.e., the cooperative construction of a model of the environment explored by the robots or mobile devices, is fundamental to geotag sensor data (e.g., for pollution monitoring, surveillance and search and rescue), and to gather situational awareness. In this work describe a design of a technique that allows each robot or mobile device to build its own object level map while asking for minimal knowledge of the map of the teammates. In particular, we present the following three major contributions:
- A distributed algorithm based on Distributed Gauss-Seidel to estimate the 3D trajectories of multiple cooperative robots from relative pose measurements. This approach has several advantages. It requires minimal information exchange, which is beneficial in presence of communication and privacy constraints. It has an anytime flavor: after few iterations, the trajectory estimates are already accurate, and they asymptotically convergence to the centralized estimate. The DGS approach scales well to large teams, is resistant to noise and it has a straightforward implementation. We test the approach in simulations and field tests, demonstrating its advantages over related techniques.
- An approach for distributed SLAM which uses object landmarks in a distributed mapping framework. We show that this approach further reduces the information exchange among robots (as compared to feature based DGS), results in a compact, human understandable map, and has lower computational complexity as compared to low-level feature based distributed mapping.
- Finally, we extend the previous work to the case where object models are previously unknown and are modeled jointly with Distributed Object-based SLAM. We show that this approach further reduces the memory required to store the object models while maintaining the accuracy at the same level as the state of art RGB-D mapping approaches.
Henrik I Christensen (IEEE/AAAS Fellow)
Qualcomm Chancellor’s Chair, UC San Diego, USA
Henrik Christensen is the director of the Contextual Robotics Institute and a professor of Computer Science and Engineering at UC San Diego. Prior to San Diego he was the director of robotics at Georgia Tech (2006-2016). Prior to this he was a professor of computer science at the Swedish Royal Institute of Technology 1998-2006. He was also the director of the Swedish Center for Autonomous Systems 1996-2006. During the same period, he was the founder and coordinator of the European Network of Excellence in Robotics, which involved more than 190 universities and companies across all European member states. He was an associate professor of robotics and computer vision at Aalborg University 1992-1996.
Henrik I Christensen received his first degree in Mechanical Engineering from the Technical College of Frederikshavn, 1981. He subsequently worked at MAN B&W on control systems designs. He earned M.Sc. and Ph.D. degrees in Electrical Engineering from Aalborg University, Denmark 1987 and 1989, respectively.
Dr. Christensen does research on a systems approach to sensor-based robotics. The research must have a solid theoretical foundation, an efficient implementation and be evaluated in realistic contexts. Consequently, the emphasis is on “real systems for real applications”.
The research has been published in more than 350 contributions across robotics, computer vision and artificial intelligence. The research has been recognized by numerous awards including best paper awards, the Joseph Engelberger Award (the highest honor by the robotics industry), and the Boeing Supplier of the Year Award 2011.
Dr. Christensen is a fellow of American Association for Advancement of Science (AAAS) and Institute of Electrical and Electronic Engineers (IEEE). He received an honorary doctorate from Aalborg University 2015. Dr. Christensen was the coordinator of the formulation of the US National Roadmaps for Robotics 2009, 2013 and 2016. The roadmaps were presented to the US Congress. He has graduated 29 PhD students and more than 60 M.Sc. students that today occupy positions at universities and companies across 3 continents.
Dr. Christensen is the co-founder of five companies and he currently serves on the board of Blue-Ocean Robotics and Robo Global. He also serves as a consultant to a number of companies and agencies across 3 continents.