Computing Danger Zones for Provably Safe Closely Spaced Parallel Approaches

Abstract

The use of a differential game and an optimal control theoretic approach for computing danger zones for this particular problem is new. The space outside the danger zone is deemed provably safe, implying that during blunder situations, emergency evasive maneuvers starting from outside the zone are guaranteed to be safe within limits on the maneuver severity of both the blundering and evading aircraft. In addition, these zones can be computed online in real time with knowledge of the current state. The algorithm also provides the minimum runway spacings for independent approaches and the minimum longitudinal separation between aircraft for dependent approaches. The algorithm for computing the zones has been validatedwith over one million simulationruns, where the adjacent aircraft performed various blunder trajectories. The validation shows that the algorithm could successfully alert and prevent collisions in 100% of the simulations ran. Rodney Teo is a Ph.D. candidate in the Department of Aeronautics and Astronautics Engineering at Stanford University. He received his B. Eng. in mechanical engineering from the National University of Singapore in 1990 and his M. S. in aeronautics and astronautics engineering from Stanford University in 1998. He has held positions as Project Engineer (1990-1995) and Project Manager (1996-1997) on helicopter acquisition and system integration projects in the Ministry of Defense of Singapore. His research interests are in vehicle dynamics, guidance and control, and in cooperative coordination, guidance, and control of multiple vehicles. He is a Student Member of AIAA.

Computational techniques for the verification of hybrid systems

“…Perhaps more exciting are the applications that we are currently working on: we have used reachable set computations to design safe emergency escape maneuvers for dual aircraft closely spaced parallel approaches [82], and we are currently working with Boeing St. Louis on a real-time application of our pursuit-evasion game (under the DARPA Software Enabled Control program). Also, we believe that these algorithms have great potential for human-automation interface design.…”

Automated multiple UAV flight - the Stanford DragonFly UAV Program

“…The details of the algorithm, and all of the calculations, are presented in [10], [13]. We have computed the danger zones for three different cases; in this paper we present the results for the Baseline case, which is shown in Figure 4.…”

“…We have applied this algorithm to Closely Spaced Parallel Approaches (CSPA) of aircraft into airports. The algorithm and its application to CSPA, is presented in detail in [10], [13].…”

“…They are fixed wing, 10 foot wingspan, fully automated aircraft. This platform has been a wonderful testbed to support and validate our research on applied advanced control: these two vehicles are providing the experimental testbed for single vehicle control and navigation [6], [7], and dual vehicle coordinated approach control [13], and will be used as part of a larger testbed for our research on decentralized optimization schemes for formation flying [3].…”

Flight Demonstration of Provably Safe Closely Spaced Parallel Approaches

“…It is a summary of the key results presented in Ref. 33 21 These algorithms, however, all assume some specific set of blunder trajectories. These algorithms can thus guarantee safety only for the specific set of blunder trajectories assumed.…”