Papers by Rohan Kumar Manna
RAL and ICRA@40, Jul 21, 2023
Vision-based object tracking is an essential precursor to performing autonomous aerial navigation... more Vision-based object tracking is an essential precursor to performing autonomous aerial navigation in order to avoid obstacles. Biologically inspired neuromorphic event cameras are emerging as a powerful alternative to frame-based cameras, due to their ability to asynchronously detect varying intensities (even in poor lighting conditions), high dynamic range, and robustness to motion blur. Spiking neural networks (SNNs) have gained traction for processing events asynchronously in an energy-efficient manner. On the other hand, physics-based artificial intelligence (AI) has gained prominence recently, as they enable embedding system knowledge via physical modeling inside traditional analog neural networks (ANNs). In this letter, we present an event-based physics-guided neuromorphic planner (EV-Planner) to perform obstacle avoidance using neuromorphic event cameras and physics-based AI. We consider the task of autonomous drone navigation where the mission is to detect moving gates and fly through them while avoiding a collision. We use event cameras to perform object detection using a shallow spiking neural network in an unsupervised fashion. Utilizing the physical equations of the brushless DC motors present in the drone rotors, we train a lightweight energy-aware physicsguided neural network (PgNN) with depth inputs. This predicts the optimal flight time responsible for generating near-minimum energy paths. We spawn the drone in the Gazebo simulator and implement a sensor-fused vision-to-planning neuro-symbolic framework using Robot Operating System (ROS). Simulation results for safe collision-free flight trajectories are presented with performance analysis, ablation study and potential future research directions.
ArXiv, 2021
Deep Neural Networks (DNNs) which are trained end-to-end have been successfully applied to solve ... more Deep Neural Networks (DNNs) which are trained end-to-end have been successfully applied to solve complex problems that we have not been able to solve in past decades. Autonomous driving is one of the most complex problems which is yet to be completely solved and autonomous racing adds more complexity and exciting challenges to this problem. Towards the challenge of applying end-to-end learning to autonomous racing, this paper shows results on two aspects: (1) Analyzing the relationship between the driving data used for training and the maximum speed at which the DNN can be successfully applied for predicting steering angle, (2) Neural network architecture and training methodology for learning steering and throttle without any feedback or recurrent connections.
SAE International Journal of Connected and Automated Vehicles, 2022
We present our approach to several control challenges in high-speed autonomous racing for the Ind... more We present our approach to several control challenges in high-speed autonomous racing for the Indy Autonomous Challenge (IAC). The IAC involves autonomous head-to-head racing at speeds approaching 200 mph. The autonomy system must maintain traction and stability when operating at such high speeds while also maneuvering aggressively around other competitor vehicles. One key challenge arises from limited actuator update frequency. We propose two lateral control methods to follow the desired trajectory: a cross-track error method and a pure pursuit lookahead angle method. Analysis shows that, when linearized, cross-track error and pure pursuit angle are related by a first-order system. To analyze the effect of actuator update frequency on closed-loop performance, we emulate the discrete rate as a time delay. Control parameters and gains for both controllers can be solved for using loop-shaping techniques to guarantee closed-loop bandwidth and phase margin. Bandwidth is limited by the time delay associated with the loop frequency.\del{ Results depict that the steering actuator update frequency should be as high as possible to maximize vehicle stability.} \rev{Results indicate that increasing the steering actuator update frequency allows for a higher closed-loop control bandwidth and increased tracking performance. (For example, a 50Hz control bandwidth can be achieved by a 1kHz actuator update frequency.)} Our acceleration and longitudinal control strategy maximizes acceleration performance while respecting traction and engine RPM limits by applying throttle and shifting gears based on the engine torque-speed profile to achieve desired wheel traction forces. We demonstrate our control strategies in simulation at various control loop frequencies. A higher control loop frequency allows for higher-bandwidth lateral controllers exhibiting a faster response to changes in the desired trajectory. From a rolling start in an initial acceleration test at a constant control loop frequency, our longitudinal control and shifting strategy outperforms the IAC default automatic transmission.
Recent improvements in the area of Internet of Things (IoT) has led to extensive utilization of e... more Recent improvements in the area of Internet of Things (IoT) has led to extensive utilization of embedded devices and sensors. Hence, along with utilization the need for safety and security of these devices also increases proportionately. In the last two decades, the side-channel attack (SCA) has become a massive threat to the interrelated embedded devices. Moreover, extensive research has led to the development of many different forms of SCA for extracting the secret key by utilizing the various leakage information. Lately, machine learning (ML) based models have been more effective in breaking complex encryption systems than the other types of SCA models. However, these ML or DL models require a lot of data for training that cannot be collected while attacking a device in a real-world situation. Thus, in this thesis, we try to solve this issue by proposing the new technique of leakage conversion. In this technique, we try to convert the high signal to noise ratio (SNR) power traces...
Uploads
Papers by Rohan Kumar Manna