Abstract
Formation control is essential for swarm robotics, enabling coordinated behavior in complex environments. In this paper, we introduce a novel formation control system for an indoor blimp swarm using a specialized leader-follower approach enhanced with a dynamic leader-switching mechanism. This strategy allows any blimp to take on the leader role, distributing maneuvering demands across the swarm and enhancing overall formation stability. Only the leader blimp is manually controlled by a human operator, while follower blimps use onboard monocular cameras and a laser altimeter for relative position and altitude estimation. A leader-switching scheme is proposed to assist the human operator to maintain stability of the swarm, especially when a sharp turn is performed. Experimental results confirm that the leader-switching mechanism effectively maintains stable formations and adapts to dynamic indoor environments while assisting human operator.
Introduction
Managing a swarm of miniature autonomous blimps (MABs) presents significant challenges for a single human operator. Blimps are inherently underactuated and possess limited payload capacities, precluding the use of heavy sensors like LiDAR. In traditional leader-follower configurations, followers often lose visual contact with the leader during sharp turns due to wide turning radii and restricted camera fields of view. This "formation breakage" necessitates frequent manual intervention. This research addresses these issues by proposing a decentralized control architecture that utilizes minimal onboard hardware and a dynamic leader-switching mechanism to maintain swarm cohesion in complex indoor environments.
System Design and Methodology
The proposed system relies on a lightweight sensor suite: a single monocular camera, an Inertial Measurement Unit (IMU), and a laser altimeter.
Relative Position Estimation and Control
Follower blimps estimate their relative position by visually tracking the leader. By analyzing the leader's apparent size and orientation within the monocular camera's frame, the system calculates distance and yaw. Altitude is maintained via the laser altimeter. Control is achieved through a series of PID controllers. Notably, the distance controller regulates horizontal displacement by adjusting the blimp's pitch angle, exploiting the physical coupling between tilt and forward acceleration to compensate for the blimps' slow response times.
Dynamic Leader-Switching Mechanism
The core innovation is a "relay" approach to leadership. When the swarm approaches a waypoint requiring a sharp turn, the system evaluates the relative positions of all blimps.
If a turn is directed to the left, a blimp already positioned on the left may assume the leader role. This redistribution of maneuvering demand ensures that followers remain within the visual range of the current leader, preventing formation dispersal. To manage computational constraints, heavy visual processing is offloaded to a ground-based PC, which broadcasts control signals back to the swarm.
Experimental Results
The system was evaluated through six real-world experimental test cases in an indoor setting, comparing flight with and without the leader-switching algorithm.
- Success Rate: The swarm achieved a 100% success rate in completing sharp turns with the leader-switching mechanism. In contrast, traditional leader-follower methods succeeded in only 33.3% of cases, as followers frequently lost sight of the leader.
- Formation Stability: Stability was quantified using the Area Root Mean Squared Error (RMSE) of the formation triangle. With switching enabled, the RMSE remained low (0.14 to 0.28), whereas it spiked significantly (2.06 to 2.56) without the mechanism.
- Operator Assistance: The system includes a "leader-search" mode where followers autonomously rotate to reconnect if visual contact is lost. The user interface also alerts operators to risky switches, effectively reducing the cognitive workload required for swarm management.
Conclusion and Future Work
This research demonstrates that physical limitations of underactuated robots can be mitigated through logical role-switching. The dynamic leader-switching mechanism significantly enhances the stability and maneuverability of MAB swarms. Future work will address scalability and visual occlusion in larger swarms (e.g., 10+ blimps) by implementing hierarchical following strategies where followers track other followers in a chain-like structure. Additionally, reducing communication latency in radio-congested environments remains a priority for improving system robustness.
Summary updated