In honor of National Robotics Week, the Computing Community Consortium is releasing a mid-cycle update to the US National Robotics Roadmap. The US National Robotics Roadmap is updated every four years. Since the last update in September 2020, the world has changed significantly. The COVID pandemic has slowed, there is another administration in Washington, the National Robotics Initiative has officially ended, and the political climate for international trade is constantly changing. Given all these aspects, the Computing Community Consortium (CCC) is publishing a minor revision of the roadmap before the next regular update in 2024.
In November, the CCC published a call for contributions and input from the community. In addition, a discussion session was organized at AAAI in Washington, DC on February 7th to collect input from the AI/Robotics community. All inputs have been considered in the preparation of the present document.
Over the past decade, national support for basic research in robotics has been significantly reduced. Recent developments and resulting gaps include the following:
- The recent sunset of the National Robotics Initiative (NRI) in May 2022 has resulted in a lack of cross-agency programs focused on robotics. A few agencies have their own program with limited cross coordination, but overall the emphasis has shifted more towards a National AI program.
- The NRI program has been replaced by the NSF Foundational Research in Robotics (FRR) program, which is focused on robot systems that include both computational and physical complexity. The new FRR program lacks direct support for component technologies within each of the areas of embodiment, perception, and planning. Consequently, many researchers in robotics and component disciplines do not have a natural program within NSF (CISE or ENG) to consider for their basic research.
- The Department of Defense (DoD) houses a number of robotics programs. The Army Research Lab (ARL) and Office of Naval Research (ONR) have programs, but mainly with clear mission objectives rather than curiosity-driven objectives. In addition, the number of performers is still modest.
- The National Institute of Health (NIH) also has programs with clear clinical objectives. In general, acceptance rates for NIH are very low. There is significant progress on design of medical devices, support for elderly people and for medical procedures. However, during the process from idea to a certified product/method, the risk of losing support is significant.
While the US has seen a reduction in support for robotics, Europe (EU-Horizon), China (Robotics+), South Korea (RRI), and India (Manufacturing in India) are all investing heavily in the technology. Already today, the US is falling behind other nations in terms of basic research and utilization of robot technology for next-generation manufacturing, logistics, and smart infrastructure. When the NRI was launched in 2011, the US was a top 5 consumer of robots for manufacturing and China was not in the top 10. Today China is the largest consumer of robots for manufacturing, and the US is now 7th according to the International Federation of Robotics (IFR, 2022). Without a concerted investment across basic research, translation and utilization, more ground will be lost.
This mid-cycle update to the US Robotics Roadmap identifies and explores recent megatrends and key research challenges in the robotics field, as well as, implementation considerations and recommendations to further robotics research and propel the US back to the forefront of robotics research. You can read the full report here.