The body and soul of any research lab is its portfolio of research projects. Our projects are grouped into seven topic areas corresponding to MERL's seven research groups.
Extracting meaning and building representations of visual objects and events in the world. Research addresses visual representation and analysis, including detection, classification, tracking and recognition, based on machine learning techniques; 3D reconstruction, mapping, and localization; human monitoring; and multi-modal sensor integration for navigation and interaction.
Speech & Audio
Machine learning for estimation and inference problems in speech and audio processing, including end-to-end speech recognition and enhancement, acoustic modeling and analysis, statistical dialog systems, as well as natural language understanding and adaptive multimodal interfaces.
Apply signal processing and computational methods to improve sensing and enable new paradigms. Research covers acquisition, representation, and processing of signals with an emphasis on statistical inference, estimation, and inverse problems. Additionally, study physics- and learning-based frameworks to model, sense and control physical processes and dynamical systems.
Connectivity & Information Processing
Apply connected intelligence to enhance the performance, resilience and security of communications networks. Develop fundamental machine learning techniques considering privacy, robustness and power efficiency concerns. Explore new areas of information and biosignal processing and quantum machine learning.
Control for Autonomy
Enabling autonomy in automotive, space, robotics, and mechatronics systems by research in control, estimation, motion planning, optimization and dynamics for both model-based and data-driven frameworks.
Optimization & Intelligent Robotics
Research on fundamental methods to solve decision
optimization problems, including continuous and combinatorial optimization, and sequential
decision making, as well as predictive modeling techniques. The robotics research explores
the intersection between dexterous
robotics and human robot interaction supported by fundamental research on perception,
machine learning, augmented/virtual reality, planning and reasoning of single and multi-agent systems.
Electric Systems Automation
Applied Physics for multi-physical modeling, design optimization, control and performance monitoring/diagnosis of devices, components and electric systems. Target applications include motors, magnetic-based equipment, and other electric machines.
Research multi-physical modeling methodology, develop and use models for simulation, analysis, control, and optimization of large-scale systems that manifest interactions between the thermal, fluid, electrical, and mechanical domains. Target applications include HVAC systems, factory automation and robotics.