I am a first year PhD student at the CoMMA Lab at Purdue University, advised by Dr. Zachary Kingston. My research aims to develop robotic agents that can reason about uncertainty while incrementally refining probabilistic models of motion.
To this end, I bring probabilistic programming to robotics. Probabilistic programming is an emerging paradigm for building inference kernels that automatically construct particle-based approximations of complex, multimodal posteriors. I hypothesize that by leveraging these kernels, robotic systems can generalize trajectory inference across discontinuities in dynamics and environments—scenarios where gradient-based methods alone often fail.
I completed my undergraduate and Master’s degrees at MIT, where I was a member of the Probabilistic Computing Project. Independently, I designed and implemented an experimental Rust crate for dynamic probabilistic programming called ModPPL inspired by Gen.jl’s Dynamic Modeling Language.
2026
arXiv
Faster Visuomotor Policy Learning on Action Manifolds via Riemannian MeanFlow
Visuomotor policies learn a direct map from raw sensory observations to robot action sequences. Policies based on Diffusion and Flow Matching capture the multimodal distribution over action sequences in an end-to-end manner. This expressivity comes at the cost of multi-step numerical integration of the learned vector field for action generation, which can be expensive and time-consuming, impeding fast control rates required in robotics applications. Furthermore, robot action sequences are usually defined on a smooth, differentiable manifold, requiring that the learned policy respects the intrinsic geometry of the robot’s action space. Here, we present Riemannian MeanFlow Policy (RMFP), which learns the conditioned flow map of the probability path on the robot action manifold. Our formulation employs a flow map consistency objective grounded in the data by a Riemannian Conditional Flow Matching anchor. The flow map consistency condition is stable to train and constrains the learned model to finite-time transport, which yields on-manifold action sequence generation with as few as one network function evaluation. We present results on the spherical LASA and Push-T benchmarks, on the Tool Hang and Transport tasks of the Robomimic suite, and on the Franka Kitchen task with manifold-constrained action generation, and demonstrate that RMFP attains performance competitive with prior work at a lower sampling cost. We also employ RMFP on a real-world robotic manipulation task to demonstrate fast action generation under imperfect sensor measurements in the physical world.
@misc{bukhari2026rmfp,archiveprefix={arXiv},author={Bukhari, S. Talha and Garrett, Austin and Wei, Yi and Ni, Ruiqi and Kingston, Zachary and Bera, Aniket},eprint={2609.30127},note={Under Review},primaryclass={cs.RO},title={Faster Visuomotor Policy Learning on Action Manifolds via {R}iemannian {MeanFlow}},year={2026}}
arXiv
Becoming a Fruit Ninja: Real-Time Probabilistic Kinodynamic Planning for Manipulator Projectile Interception
Projectile interception is a challenging dynamic manipulation problem. Intercepting a thrown object with a robot arm requires reaching a point on the object’s path as the object passes through it. Slicing also fixes the blade’s velocity and orientation at contact. The goal is therefore a subset of the states of the robot and arrival times that moves as the object falls, and the arm must reach it within its actuator limits in milliseconds. We present FRUITNINJA, an anytime sampling-based planner that grows a tree on the GPU in batches toward the interception manifold. Each edge is an exact cubic whose travel time is found by a parallel search against the arm’s dynamics, so every edge satisfies the actuator limits. Plans are ranked by a risk-aware objective over the uncertainty in the object’s position and the arm’s arrival time. We evaluate on a Franka Research 3 against six baselines in a calibrated real-time simulator, where FRUITNINJA cuts 96.7% of tosses in the open and 68.3% among five obstacles, versus the best baseline’s 68.3% and 35.0% respectively.
@misc{chen2026fruitninja,archiveprefix={arXiv},author={Chen, Lucas and Garrett, Austin and Niu, Andrew and Kingston, Zachary},eprint={2609.22608},note={Under Review},primaryclass={cs.RO},title={Becoming a Fruit Ninja: Real-Time Probabilistic Kinodynamic Planning for Manipulator Projectile Interception},year={2026}}
