▍The Mystery of Collective Decision-Making in Bees
The collective behavior of bees has always been a fascinating natural mystery, especially their ability to make collective decisions when choosing new hive locations, which is truly remarkable.When a hive becomes unsuitable for habitation for some reason, the swarm sends out scout bees to survey potential new hive sites in the surrounding area. These scout bees observe and compare the quality, size, and other conditions of different locations, then return to the swarm to communicate with other bees through the famous “waggle dance,” gradually forming a collective consensus on the best candidate site.
Just like elections in human society, different individuals make choices based on their judgments, and then through discussion and debate, they gradually reach a consensus, determining the optimal solution from many alternatives.The entire decision-making process is completely decentralized and self-organized, with no “command center,” yet it can still democratically determine the location of a new hive.
For many years, this collective decision-making mechanism in bees has sparked strong curiosity and interest among scientists. To unveil its mysteries, researchers have established many mathematical models to simulate the selection process of bees and have derived some interesting theoretical results. However, these models are often based on the assumption of “complete connectivity,” meaning that every bee can communicate with all other members in the group. In reality, the communication between bees is clearly limited to a certain range. So, what impact does local interaction have on the collective decision-making results of bees? Especially when bees can move in space, how does this movement change their communication structure?These questions have greatly intrigued researchers.
▍Miniature Robots Simulating Bee Swarms
To explore the impact of spatial constraints and local interactions on collective decision-making, a group of researchers adopted a very innovative research approach—using miniature robots to simulate the collective decision-making process of bees!
These miniature robots, called “kilobots,” are only 3.3 cm long, comparable to the size of a real bee.They can autonomously move within a confined circular area and display their “choices” using LED lights. More importantly, these robots can communicate information over short distances using infrared communication, just like real bees communicate through their dances.
The researchers placed 35 kilobots in a circular experimental area with a radius of 20 cm, set up two virtual candidate hive locations (Location 1 and Location 2), and assigned them different “quality values” (Location 1 has a quality value of 7, Location 2 has a quality value of 10). Then, they wrote code to guide the behaviors of these miniature robotic bees based on the rules of the bee hive selection model.
In the model, each bee is an independent individual that can autonomously discover different candidate locations and can also obtain information about these locations by observing the dances of other bees. The probability of a location being discovered is its “spontaneous discovery probability,” while the extent to which a bee observes and is influenced by the dances of other bees is called “interdependence.” Researchers can adjust these two parameters to study the nature of collective decision-making.
Like real bees, these robotic bees can autonomously move under program control, and when they discover a location, they will “dance” for that site and influence other robotic bees. Meanwhile, undecided robotic bees will stop to “observe” the dances of other bees. After a period of interaction, the entire swarm of robotic bees will eventually form a collective consensus on one of the candidate locations.
▍The Power of Movement, the Magic of Communication
By observing the interactions of these mini robots, researchers discovered some very interesting phenomena.
(1) The Power of Movement
When the number of robotic bees is small, their communication is very limited, and the final selection result is quite random. However, as the number of bees increases, the movement trajectories of the robotic bees start to overlap and intertwine more, which greatly facilitates the spread of information within the group. Under sufficiently crowded conditions, the selection results of these robotic bees are as good as those in an “ideal fully connected” situation, successfully reaching a consensus on the higher quality location 2!
It turns out that when robotic bees can move within the area, their movement trajectories act like a large network, connecting seemingly isolated individuals. Information is then carried to all corners by this “moving airflow” like water molecules in the air. It can be said that movement itself becomes a “catalyst” for collective decision-making, greatly facilitating communication between seemingly isolated individuals.
This discovery has important implications for human society as well: each of us can expand our vision and communicate with more people through our actions. Just as “movement” is a prerequisite for communication, proactive actions can eliminate apparent barriers and connect more possible perspectives.
(2) The Magic of Communication
Research also found that when the “interdependence” among robotic bees, i.e., the proportion of mutual observation, is low, they cannot reach a clear consensus, and individual choices are highly random. However, as “interdependence” increases, meaning they rely more on their peers’ opinions, the consensus formed is clearer, and the quality of the chosen hive site is higher.
This is very similar to human society. If every individual is very stubborn and does not listen to any other opinions, it is very difficult to form correct collective decisions. Only by appropriately referencing others’ viewpoints and leveraging collective wisdom can we overcome individual limitations and achieve higher-quality conclusions.
The magic of communication can be said to stem from its ability to integrate different perspectives, forming a more comprehensive understanding. Repeating the exact same viewpoint multiple times does not bring new information, while complementary different viewpoints can broaden our horizons.
▍Inspiration and Future Prospects
This research utilizing miniature robots to simulate bee collective decision-making greatly expands our understanding of the laws of self-organizing systems. It proves that seemingly simple and relatively isolated individuals can achieve complex collective decision-making through local interactions and movement. This provides valuable inspiration for further improving the collective behavior of artificial intelligence systems.
Of course, this is just the beginning; there are many directions worth exploring in the future. For example, we can continue to increase the number of robots and observe the impact of group size on decision-making results; set up more candidate choices to study the scalability of collective decision-making; add more noise among the robots to simulate the complexity of the real world, and so on.
On the other hand, this research also makes us marvel at the wonders of biological organisms in nature and the laws hidden within waiting for us to discover. These seemingly weak little creatures, through complex interactions formed by evolution, can produce immense collective wisdom.
From the microscopic bees to the macroscopic human society, understanding the mechanisms of collective decision-making will continue to inspire endless scientific exploration and technological innovation. Let us continue to observe, simulate, and optimize various collective systems, including humans, with an open and truth-seeking heart!
—————–END——————-
More Exciting Content
* Q4 this year will be mass-produced! Yushu released the H1 universal humanoid robot equipped with high-torque and high-explosion M107 joint motors
* In one day, two companies announced their entry into the humanoid robot track, and the capital market faced a contrasting situation
* Is it an IQ tax or a real necessity? What is the actual use of the laser radar that the industry is hotly discussing?
* The industry’s first omnidirectional ultra-wide-angle 4D laser radar is here!
* Important progress! Zhejiang University develops a self-protecting electrohydraulic soft robot published in Nature sub-journal
* Can fly without a battery? The “smart leaf” autonomously changes shape in the air, published in Science sub-journal, only half the weight of a nail
* The first robot hip and knee replacement surgery in southern Xinjiang! Domestic surgical robots break the dilemma of resource allocation in medical advantages
* The “new internet celebrity” at the Asian Para Games, Yushu Technology’s quadruped robot B1 guiding service, industry innovation upgrades again
* Revenue growth of 35.38%, Estun’s market share is expected to continue to rise!
* Robots become “point-stepping maniacs”! Following the beat freestyle, the University of Bremen has open-sourced
* The impact of NVIDIA’s high-performance chip supply interruption on the robot industry
* Truly deserving! Congratulations to Professor Gao Huijun from Harbin Institute of Technology for receiving the highest award in the field of systems and control theory!
* Zhejiang University FAST-Lab’s Gao Fei team announces new research progress at IROS 2023
* Dynamic perception algorithms + IoT, this company has further “upgraded” its technology
* Breaking news! Zhi Ji Power has raised nearly 200 million yuan in financing! The market is optimistic about the application prospects of general-purpose biped robots
* -20°C to 55°C extreme environments undergo practical testing, this quadruped robot is gaining popularity!
* Really astonishing! How does a 130g soft gripper manage to lift a 100kg object?
* Able to steadily lift a 27kg dumbbell above, while gently pinching ash or catching goldfish below, Huazhong University of Science and Technology’s rigid-flexible coupling gripper showcases super strong performance!
* The localization process of smart sensors is accelerating, and another company is ramping up its entry into the industrial field
* Benmo Technology’s wheeled robot is entering a 57.1 billion yuan inspection market, challenging a 14.5% industry growth rate
Join the Community
Welcome to join the 【Robot Lecture Hall】 reader discussion group, to discuss topics related to the field of robotics and share cutting-edge technology and industry dynamics.
Educational robots, medical robots, legged robots, industrial robots, service robots, special robots, drones, soft robots, and other professional discussion groups are recruiting, follow the Robot Lecture Hall public account, send ” Discussion Group ” to get the joining method!
Recruiting Authors
Robot Lecture Hall is recruiting 【part-time content creators】. If you are interested in writing articles related to robotics 【technology】 or 【industry】, please send your resume and original works to the email: [email protected]
We have no requirements for profession or location, and welcome friends to join!
Are you tired? Tap “View” to support us!