BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Department of Aerospace Engineering - ECPv6.6.3//NONSGML v1.0//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
X-ORIGINAL-URL:https://aero.iisc.ac.in
X-WR-CALDESC:Events for Department of Aerospace Engineering
REFRESH-INTERVAL;VALUE=DURATION:PT1H
X-Robots-Tag:noindex
X-PUBLISHED-TTL:PT1H
BEGIN:VTIMEZONE
TZID:Asia/Kolkata
BEGIN:STANDARD
TZOFFSETFROM:+0530
TZOFFSETTO:+0530
TZNAME:IST
DTSTART:20260101T000000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTART;TZID=Asia/Kolkata:20260209T111500
DTEND;TZID=Asia/Kolkata:20260209T130000
DTSTAMP:20260521T110555
CREATED:20260202T091939Z
LAST-MODIFIED:20260204T061202Z
UID:10000113-1770635700-1770642000@aero.iisc.ac.in
SUMMARY:Ph.D. (Engg) : Multi-Agent Coordination using Convex Formations and Binary Tree Structures
DESCRIPTION:Multi-agent systems are increasingly deployed in missions involving large-scale tasks with complex objectives that are beyond the capability of a single agent. Such missions demand computationally efficient coordination strategies that ensure safety\, reliable operation\, and ease of implementation\, particularly in dynamic and uncertain environments. This thesis investigates coordination strategies in multi-agent systems\, specifically addressing the problems of distribution of agents on an enclosing boundary\, cooperative target capture and containment\, and traversal through constrained spaces.\n\nThe first part of the thesis presents a convex layer-based strategy that assigns collision-free paths to a swarm of point-sized agents to reach an enclosing circular boundary. Leveraging the construction of convex layers from the initial positions of agents\, a novel search space for an agent on a convex layer is defined as an angular region enclosed between the lines passing through the agent’s position and normal to its supporting edges. A goal assignment policy is proposed\, which designates a unique goal position on the boundary within the search space of an agent. Subsequently\, the proposed framework is extended to polygonal boundaries\, considering disc-shaped agents. Therein\, the proposed policy assigns a goal position to each agent in order of decreasing overlap between their search spaces and the polygonal boundary\, while excluding angular regions corresponding to already assigned goal positions. Further\, a layer-wise speed assignment rule is proposed\, which ensures collision-free trajectories for the agents. Simulation studies assess the proposed method under various real-world considerations\, including the finite size of the agents\, a six-degree-of-freedom quadrotor model\, uncertainties in initial position information\, and communication delays.\n\nIn the second part\, the problem of multiple pursuers engaging a single evader is considered in two complementary scenarios. Firstly\, the problem of capturing the evader in an unbounded region is addressed. As the key construct\, the evader’s proximity region is characterized by the region generated by the Voronoi diagram constructed using the positions of the pursuers and the evader. Pursuers’ velocity inputs are deduced as a function of the position and velocity of the vertices of the evader’s proximity region and the evader. A motion policy is proposed that directs the vertices of the evader’s proximity region toward its centroid\, under which the region is analytically shown to shrink exponentially over time\, irrespective of the evader’s motion policy. In addition\, using the Chebyshev radius of the proximity region\, an upper bound on the time of evader capture is derived. Simulation studies demonstrate the effectiveness of the proposed method under various evader maneuvers and in scenarios where evader position information is noisy. In a scenario complementary to evader capture\, a containment problem is considered\, wherein multiple pursuers are desired to encapsulate a moving evader. Considering the engagement between the evader and the centroid of the convex hull of pursuers\, a variable deviated pursuit guidance law is proposed\, which achieves a tail-chase rendezvous between the evader and the centroid. Subsequently\, a cooperative control strategy is presented\, which drives the convex hull of pursuers to confine the evader through a prescribed edge while preserving the formation rigidity. Simulation results demonstrate the efficacy of the proposed method under various evader maneuvers.\n\nThe final part of the thesis addresses the problem of sequential traversal of multiple UAVs through a narrow gap. A hierarchical binary tree is constructed with its nodes defined by the UAVs’ initial positions and the gap entry point\, presenting a routing framework that provides an ordered sequence of waypoints to each UAV. A cost function is formulated that accounts for the UAV path lengths and the angles between branches at the tree nodes\, and a binary tree is constructed by minimizing that cost using a genetic algorithm coupled with a greedy strategy. In conjunction\, a decentralized scheduling policy is proposed\, in which each UAV is assigned conflict-free time slots at nodes that are identified with potential collisions. Simulation scenarios illustrate the effectiveness of the proposed method\, and Monte Carlo studies assess its scalability.\n\nOverall\, the thesis presents deterministic and computationally efficient multi-agent coordination strategies by leveraging ideas from convex geometry and binary trees. Experimental flight trials on a nano-quadrotor platform are also conducted\, further demonstrating the practicality of the proposed coordination methods.\n\nSpeaker : Gautam Kumar \n\nResearch Supervisor : Ashwini Ratnoo
URL:https://aero.iisc.ac.in/event/ph-d-engg-multi-agent-coordination-using-convex-formations-and-binary-tree-structures/
LOCATION:STC Seminar Hall\, Dept. of Aerospace Engineering
CATEGORIES:Thesis Colloquium / Defence
ATTACH;FMTTYPE=image/jpeg:https://aero.iisc.ac.in/wp-content/uploads/2026/02/Gautam.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Asia/Kolkata:20260217T110000
DTEND;TZID=Asia/Kolkata:20260217T130000
DTSTAMP:20260521T110555
CREATED:20260213T055505Z
LAST-MODIFIED:20260213T055505Z
UID:10000115-1771326000-1771333200@aero.iisc.ac.in
SUMMARY:Ph.D. (Engg) :Experimental Investigation of Autoignition Pathways and Shock-Train Dynamics During Mode Transition in a Dual-Mode Supersonic Cavity Combustor
DESCRIPTION:Hypersonic propulsion systems capable of sustained atmospheric flight are critical enablers for future reusable launch vehicles\, long-range high-speed transport\, and responsive global strike platforms. Among the various air-breathing concepts\, scramjet engines offer unmatched efficiency at hypersonic speeds by utilizing atmospheric oxygen and avoiding the mass penalties associated with onboard oxidizers. However\, the practical realization of scramjet propulsion is fundamentally constrained by two interrelated challenges: reliable ignition and flame stabilization under extremely short residence times\, and robust operation across a wide flight envelope that necessitates smooth transition between supersonic (scramjet) and subsonic (ramjet) combustion modes. Dual-mode scramjets (DMSJ) are designed to address this requirement\, but their operability is limited by complex\, strongly coupled interactions between shock structures\, boundary-layer separation\, fuel-air mixing\, chemical kinetics\, and unsteady pressure fields during mode transition. A central difficulty in hypersonic combustors is that global flow conditions typically yield Damköhler numbers well below unity\, rendering conventional flame-holding ineffective. Localized enhancement of thermochemical coupling through elevated temperature\, pressure\, and residence time is therefore essential to initiate and sustain combustion. Cavity-based flameholders have emerged as a promising solution due to their passive\, low-drag configuration and ability to generate recirculation zones that promote autoignition and flame anchoring. Nevertheless\, cavity-stabilized combustors introduce additional challenges: strong sensitivity to geometry\, concentration of thermal loads\, susceptibility to unsteady shear-layer oscillations\, and complex coupling with shock-train dynamics during scram-to-ram transition. Despite extensive cold-flow investigations of isolator shock trains\, their behaviour under reacting\, high-enthalpy conditions where heat release actively modifies the flow remains insufficiently characterized. This doctoral research discusses a systematic experimental investigation of autoignition pathways\, flame stabilization mechanisms\, and shock-train dynamics in a cavity-stabilized dual-mode supersonic combustor. Experiments are conducted in a direct-connect high-enthalpy facility at the Advanced Propulsion Research Laboratory (APRL)\, Indian Institute of Science. The combustor operates at flight relevant conditions of total temperature of 1500 ± 30 K and static pressure of 43 kPa\, which corresponds to Mach 5.5 flight conditions at 28 km altitude. The experimental test article features an optically accessible supersonic combustor with a single/twin cavity configuration and is designed for an inlet Mach 2.5. Time-resolved Schlieren imaging\, CH* and C2* chemiluminescence\, and high-frequency wall-pressure measurements are employed to resolve unsteady flow-flame interactions governing ignition and mode transition. Two cavity geometries with identical depth but different length-to-height ratios (L/H = 5 and 8.5) were examined to quantify the influence of geometry on ignition robustness and shock–flame coupling. For the L/H = 5 configuration\, ethylene ignition occurred downstream in the diverging duct at a global equivalence ratio of ϕg ≈ 0.3\, followed by upstream flame propagation and eventual stabilization along the shear layer. In contrast\, the L/H = 8.5 cavity enabled earlier and more robust ignition upstream\, triggered by shock-assisted autoignition behind an X-type shock formed through interaction between the cavity reattachment shock and a top-wall separation bubble. The larger cavity generated stronger pressure deficits\, deeper shear-layer penetration\, and self-sustained oscillations at approximately 527 Hz\, highlighting the critical role of cavity geometry in enhancing local Damköhler numbers. Optical diagnostics technique of two-wavelength chemiluminescence (CH* and C2*) revealed ignition kernels forming preferentially in high-temperature lean regions before stabilizing near stoichiometric zones. Shock-induced compression was shown to significantly reduce ignition delay\, enabling autoignition even for fuels with substantially longer chemical timescales. Fuel-blending experiments established a limiting ignition-delay threshold\, providing quantitative guidance for fuel selection in practical hypersonic combustors. The scram-to-ram mode transition occurred at ϕg ≈ 0.58 for both geometries and was marked by the formation of a pre-combustion shock train\, initiated due to combustion induced boundary layer separation. The L/H = 8.5 cavity sustained stable ram-mode operation\, whereas the L/H = 5 configuration frequently reverted to early scram-mode behavior\, indicating weaker shock-flame coupling and reduced buffering capacity against back-pressure fluctuations. Scaling analysis of shock-train dynamics yielded Strouhal numbers (St) an order of magnitude lower than the reported values in the literature based on isothermal shock-train oscillation studies. This demonstrated the dominant influence of heat release and shock-train coupling. Proper orthogonal decomposition (POD) analysis further revealed tight coupling between shock-train motion and upstream flame propagation\, identifying critical regions in the combustor with substantial heat release fluctuations. Finally\, symmetric dual-cavity configurations were explored to assess coupled shear-layer dynamics. While dual cavities enhance residence time\, their interaction introduces additional unsteady modes\, underscoring the need for geom etry-aware stabilization strategies. Overall\, this work directly addresses critical propulsion challenges for hypersonic vehicles by elucidating the mechanisms governing ignition reliability\, shock-assisted autoignition\, and mode-transition stability in cavity-based dual-mode scramjets. The findings provide mechanistic understanding and scalable design guidelines essential for the development of robust\, operable hypersonic air-breathing propulsion systems. \n  \nSpeaker :  Sumit Lonkar \nResearch Supervisor: Pratikash Prakash Panda
URL:https://aero.iisc.ac.in/event/ph-d-engg-experimental-investigation-of-autoignition-pathways-and-shock-train-dynamics-during-mode-transition-in-a-dual-mode-supersonic-cavity-combustor/
LOCATION:STC Seminar Hall\, Dept. of Aerospace Engineering
CATEGORIES:Thesis Colloquium / Defence
ATTACH;FMTTYPE=image/jpeg:https://aero.iisc.ac.in/wp-content/uploads/2026/02/SUMIT.jpg
END:VEVENT
END:VCALENDAR