Accurate system identification is crucial for reducing trajectory drift in bipedal locomotion, particularly in reinforcement learning and model-based control. In this paper, we present a novel control framework that integrates system identification into the reinforcement learning training loop using differentiable simulation. Unlike traditional approaches that rely on direct torque measurements, our method estimates system parameters using only trajectory data (positions, velocities) and control inputs. We leverage the differentiable simulator MuJoCo-XLA to optimize system parameters, ensuring that simulated robot behavior closely aligns with real-world motion. This framework enables scalable and flexible parameter optimization. Accurate system identification is crucial for reducing trajectory drift in bipedal locomotion, particularly in reinforcement learning and model-based control. In this paper, we present a novel control framework that integrates system identification into the reinforcement learning training loop using differentiable simulation. Unlike traditional approaches that rely on direct torque measurements, our method estimates system parameters using only trajectory data (positions, velocities) and control inputs. We leverage the differentiable simulator MuJoCo-XLA to optimize system parameters, ensuring that simulated robot behavior closely aligns with real-world motion. This framework enables scalable and flexible parameter optimization. It supports fundamental physical properties such as mass and inertia. Additionally, it handles complex system nonlinear behaviors, including advanced friction models, through neural network approximations. Experimental results show that our framework significantly improves trajectory following.

Processes of evidence accumulation for motor control contribute to the ecological validity of driver models. According to established theories of cognition, drivers make control adjustments when a process of accumulation of perceptual inputs reaches a decision boundary. Unfortunately, there is not a standard way for building such models, limiting their use. Current implementations are hand-crafted, lack adaptability, and rely on inefficient optimization techniques that do not scale well with large datasets. This paper introduces Akkumula, an evidence accumulation modelling framework built using deep learning techniques to leverage established coding libraries, gradient optimization, and large batch training. The core of the library is based on Spiking Neural Networks, whose operation mimic the evidence accumulation process in the biological brain. The model was tested on data collected during a test-track experiment. Results are promising. The model fits well the time course of vehicle control (brake, accelerate, steering) based on vehicle sensor data. The perceptual inputs are extracted by a dedicated neural network, increasing the context-awareness of the model in dynamic scenarios. Akkumula integrates with existing machine learning architectures, benefits from continuous advancements in deep learning, efficiently processes large datasets, adapts to diverse driving scenarios, and maintains a degree of transparency in its core mechanisms.

Instead, supervision can be available In this work, we tackle the problem of motor in the shape of desired sensory observations, for instance sequence learning for an embodied agent. A provided by a teaching agent. In the case wide range of approaches have been proposed of handwriting, these desired sensory observations to model sequential data, using various types of are visual observations of the target letters. In reinforcement neural architectures (Recurrent Neural Networks learning, the preference for certain sensory (RNNs), Long Short-Term Memories (LSTMs) states is modeled by assigning rewards to the [1], Restricted Boltzmann Machines (RBMs) [2]) desired states, and the agent learns a behavioral and various learning strategies (backpropagation policy maximizing its expected return (sum of rewards) through time (BPTT), Real-Time Recurrent over time. Alternatively, Active Inference Learning (RTRL) [3], Reservoir Computing (RC) (AIF) [6, 7], derived from the Free Energy Principle [4, 5]).

In this article, we apply the Free-Energy Principle to the question of motor primitives learning. An echo-state network is used to generate motor trajectories. We combine this network with a perception module and a controller that can influence its dynamics. This new compound network permits the autonomous learning of a repertoire of motor trajectories. To evaluate the repertoires built with our method, we exploit them in a handwriting task where primitives are chained to produce long-range sequences.